⛑️ feat: Helm serviceAccount Config (#9606)

* ✨ v0.8.0-rc4

* chore: update jest.config.cjs to include release comment and linting

* chore: bump CONFIG_VERSION to 1.2.9

.env.example

@@ -15,6 +15,20 @@ HOST=localhost
 PORT=3080
 
 MONGO_URI=mongodb://127.0.0.1:27017/LibreChat
+#The maximum number of connections in the connection pool. */
+MONGO_MAX_POOL_SIZE=
+#The minimum number of connections in the connection pool. */
+MONGO_MIN_POOL_SIZE=
+#The maximum number of connections that may be in the process of being established concurrently by the connection pool. */
+MONGO_MAX_CONNECTING=
+#The maximum number of milliseconds that a connection can remain idle in the pool before being removed and closed. */
+MONGO_MAX_IDLE_TIME_MS=
+#The maximum time in milliseconds that a thread can wait for a connection to become available. */
+MONGO_WAIT_QUEUE_TIMEOUT_MS=
+# Set to false to disable automatic index creation for all models associated with this connection. */
+MONGO_AUTO_INDEX=
+# Set to `false` to disable Mongoose automatically calling `createCollection()` on every model created on this connection. */
+MONGO_AUTO_CREATE=
 
 DOMAIN_CLIENT=http://localhost:3080
 DOMAIN_SERVER=http://localhost:3080
@@ -26,6 +40,13 @@ NO_INDEX=true
 # Defaulted to 1.
 TRUST_PROXY=1
 
+# Minimum password length for user authentication
+# Default: 8
+# Note: When using LDAP authentication, you may want to set this to 1 
+# to bypass local password validation, as LDAP servers handle their own
+# password policies.
+# MIN_PASSWORD_LENGTH=8
+
 #===============#
 # JSON Logging  #
 #===============#
@@ -58,7 +79,7 @@ DEBUG_CONSOLE=false
 #                     Endpoints                     #
 #===================================================#
 
-# ENDPOINTS=openAI,assistants,azureOpenAI,google,gptPlugins,anthropic
+# ENDPOINTS=openAI,assistants,azureOpenAI,google,anthropic
 
 PROXY=
 
@@ -142,10 +163,10 @@ GOOGLE_KEY=user_provided
 # GOOGLE_AUTH_HEADER=true
 
 # Gemini API (AI Studio)
-# GOOGLE_MODELS=gemini-2.5-pro-preview-05-06,gemini-2.5-flash-preview-04-17,gemini-2.0-flash-001,gemini-2.0-flash-exp,gemini-2.0-flash-lite-001,gemini-1.5-pro-002,gemini-1.5-flash-002
+# GOOGLE_MODELS=gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite-preview-06-17,gemini-2.0-flash,gemini-2.0-flash-lite
 
 # Vertex AI
-# GOOGLE_MODELS=gemini-2.5-pro-preview-05-06,gemini-2.5-flash-preview-04-17,gemini-2.0-flash-001,gemini-2.0-flash-exp,gemini-2.0-flash-lite-001,gemini-1.5-pro-002,gemini-1.5-flash-002
+# GOOGLE_MODELS=gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite-preview-06-17,gemini-2.0-flash-001,gemini-2.0-flash-lite-001
 
 # GOOGLE_TITLE_MODEL=gemini-2.0-flash-lite-001
 
@@ -349,6 +370,11 @@ REGISTRATION_VIOLATION_SCORE=1
 CONCURRENT_VIOLATION_SCORE=1
 MESSAGE_VIOLATION_SCORE=1
 NON_BROWSER_VIOLATION_SCORE=20
+TTS_VIOLATION_SCORE=0
+STT_VIOLATION_SCORE=0
+FORK_VIOLATION_SCORE=0
+IMPORT_VIOLATION_SCORE=0
+FILE_UPLOAD_VIOLATION_SCORE=0
 
 LOGIN_MAX=7
 LOGIN_WINDOW=5
@@ -437,6 +463,8 @@ OPENID_REQUIRED_ROLE_PARAMETER_PATH=
 OPENID_USERNAME_CLAIM=
 # Set to determine which user info property returned from OpenID Provider to store as the User's name
 OPENID_NAME_CLAIM=
+# Optional audience parameter for OpenID authorization requests
+OPENID_AUDIENCE=
 
 OPENID_BUTTON_LABEL=
 OPENID_IMAGE_URL=
@@ -453,11 +481,26 @@ OPENID_REUSE_TOKENS=
 OPENID_JWKS_URL_CACHE_ENABLED=
 OPENID_JWKS_URL_CACHE_TIME= # 600000 ms eq to 10 minutes leave empty to disable caching
 #Set to true to trigger token exchange flow to acquire access token for the userinfo endpoint.
-OPENID_ON_BEHALF_FLOW_FOR_USERINFRO_REQUIRED=
-OPENID_ON_BEHALF_FLOW_USERINFRO_SCOPE = "user.read" # example for Scope Needed for Microsoft Graph API
+OPENID_ON_BEHALF_FLOW_FOR_USERINFO_REQUIRED=
+OPENID_ON_BEHALF_FLOW_USERINFO_SCOPE="user.read" # example for Scope Needed for Microsoft Graph API
 # Set to true to use the OpenID Connect end session endpoint for logout
 OPENID_USE_END_SESSION_ENDPOINT=
 
+#========================#
+# SharePoint Integration #
+#========================#
+# Requires Entra ID (OpenID) authentication to be configured
+
+# Enable SharePoint file picker in chat and agent panels
+# ENABLE_SHAREPOINT_FILEPICKER=true
+
+# SharePoint tenant base URL (e.g., https://yourtenant.sharepoint.com)
+# SHAREPOINT_BASE_URL=https://yourtenant.sharepoint.com
+
+# Microsoft Graph API And SharePoint scopes for file picker
+# SHAREPOINT_PICKER_SHAREPOINT_SCOPE==https://yourtenant.sharepoint.com/AllSites.Read
+# SHAREPOINT_PICKER_GRAPH_SCOPE=Files.Read.All
+#========================#
 
 # SAML
 # Note: If OpenID is enabled, SAML authentication will be automatically disabled.
@@ -485,6 +528,21 @@ SAML_IMAGE_URL=
 # SAML_USE_AUTHN_RESPONSE_SIGNED=
 
 
+#===============================================#
+# Microsoft Graph API / Entra ID Integration  #
+#===============================================#
+
+# Enable Entra ID people search integration in permissions/sharing system
+# When enabled, the people picker will search both local database and Entra ID
+USE_ENTRA_ID_FOR_PEOPLE_SEARCH=false
+
+# When enabled, entra id groups owners will be considered as members of the group
+ENTRA_ID_INCLUDE_OWNERS_AS_MEMBERS=false
+
+# Microsoft Graph API scopes needed for people/group search
+# Default scopes provide access to user profiles and group memberships
+OPENID_GRAPH_SCOPES=User.Read,People.Read,GroupMember.Read.All
+
 # LDAP
 LDAP_URL=
 LDAP_BIND_DN=
@@ -575,6 +633,10 @@ ALLOW_SHARED_LINKS_PUBLIC=true
 # If you have another service in front of your LibreChat doing compression, disable express based compression here
 # DISABLE_COMPRESSION=true
 
+# If you have gzipped version of uploaded image images in the same folder, this will enable gzip scan and serving of these images
+# Note: The images folder will be scanned on startup and a ma kept in memory. Be careful for large number of images.
+# ENABLE_IMAGE_OUTPUT_GZIP_SCAN=true
+
 #===================================================#
 #                        UI                         #
 #===================================================#
@@ -592,11 +654,44 @@ HELP_AND_FAQ_URL=https://librechat.ai
 # REDIS Options #
 #===============#
 
-# REDIS_URI=10.10.10.10:6379
+# Enable Redis for caching and session storage
 # USE_REDIS=true
 
-# USE_REDIS_CLUSTER=true
-# REDIS_CA=/path/to/ca.crt
+# Single Redis instance
+# REDIS_URI=redis://127.0.0.1:6379
+
+# Redis cluster (multiple nodes)
+# REDIS_URI=redis://127.0.0.1:7001,redis://127.0.0.1:7002,redis://127.0.0.1:7003
+
+# Redis with TLS/SSL encryption and CA certificate
+# REDIS_URI=rediss://127.0.0.1:6380
+# REDIS_CA=/path/to/ca-cert.pem
+
+# Elasticache may need to use an alternate dnsLookup for TLS connections.  see "Special Note: Aws Elasticache Clusters with TLS" on this webpage: https://www.npmjs.com/package/ioredis
+# Enable alternative dnsLookup for redis
+# REDIS_USE_ALTERNATIVE_DNS_LOOKUP=true
+
+# Redis authentication (if required)
+# REDIS_USERNAME=your_redis_username
+# REDIS_PASSWORD=your_redis_password
+
+# Redis key prefix configuration
+# Use environment variable name for dynamic prefix (recommended for cloud deployments)
+# REDIS_KEY_PREFIX_VAR=K_REVISION
+# Or use static prefix directly
+# REDIS_KEY_PREFIX=librechat
+
+# Redis connection limits
+# REDIS_MAX_LISTENERS=40
+
+# Redis ping interval in seconds (0 = disabled, >0 = enabled)
+# When set to a positive integer, Redis clients will ping the server at this interval to keep connections alive
+# When unset or 0, no pinging is performed (recommended for most use cases)
+# REDIS_PING_INTERVAL=300
+
+# Force specific cache namespaces to use in-memory storage even when Redis is enabled
+# Comma-separated list of CacheKeys (e.g., ROLES,MESSAGES)
+# FORCED_IN_MEMORY_CACHE_NAMESPACES=ROLES,MESSAGES
 
 #==================================================#
 #                      Others                      #
@@ -657,4 +752,17 @@ OPENWEATHER_API_KEY=
 # Reranker (Required)
 # JINA_API_KEY=your_jina_api_key
 # or
-# COHERE_API_KEY=your_cohere_api_key
+# COHERE_API_KEY=your_cohere_api_key
+
+#======================#
+# MCP Configuration    #
+#======================#
+
+# Treat 401/403 responses as OAuth requirement when no oauth metadata found
+# MCP_OAUTH_ON_AUTH_ERROR=true
+
+# Timeout for OAuth detection requests in milliseconds
+# MCP_OAUTH_DETECTION_TIMEOUT=5000
+
+# Cache connection status checks for this many milliseconds to avoid expensive verification
+# MCP_CONNECTION_CHECK_TTL=60000

.github/CONTRIBUTING.md

@@ -147,7 +147,7 @@ Apply the following naming conventions to branches, labels, and other Git-relate
 ## 8. Module Import Conventions
 
 - `npm` packages first, 
-     - from shortest line (top) to longest (bottom)
+     - from longest line (top) to shortest (bottom)
 
 - Followed by typescript types (pertains to data-provider and client workspaces)
      - longest line (top) to shortest (bottom)
@@ -157,6 +157,8 @@ Apply the following naming conventions to branches, labels, and other Git-relate
      - longest line (top) to shortest (bottom)
      - imports with alias `~` treated the same as relative import with respect to line length
 
+**Note:** ESLint will automatically enforce these import conventions when you run `npm run lint --fix` or through pre-commit hooks.
+
 ---
 
 Please ensure that you adapt this summary to fit the specific context and nuances of your project.

.github/workflows/backend-review.yml

@@ -7,7 +7,7 @@ on:
       - release/*
     paths:
       - 'api/**'
-      - 'packages/api/**'
+      - 'packages/**'
 jobs:
   tests_Backend:
     name: Run Backend unit tests

.github/workflows/client.yml

@@ -0,0 +1,58 @@
+name: Publish `@librechat/client` to NPM
+
+on:
+  push:
+    branches:
+      - main
+    paths:
+      - 'packages/client/package.json'
+  workflow_dispatch:
+    inputs:
+      reason:
+        description: 'Reason for manual trigger'
+        required: false
+        default: 'Manual publish requested'
+
+jobs:
+  build-and-publish:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      
+      - name: Use Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20.x'
+        
+      - name: Install client dependencies
+        run: cd packages/client && npm ci
+        
+      - name: Build client
+        run: cd packages/client && npm run build
+        
+      - name: Set up npm authentication
+        run: echo "//registry.npmjs.org/:_authToken=${{ secrets.PUBLISH_NPM_TOKEN }}" > ~/.npmrc
+        
+      - name: Check version change
+        id: check
+        working-directory: packages/client
+        run: |
+          PACKAGE_VERSION=$(node -p "require('./package.json').version")
+          PUBLISHED_VERSION=$(npm view @librechat/client version 2>/dev/null || echo "0.0.0")
+          if [ "$PACKAGE_VERSION" = "$PUBLISHED_VERSION" ]; then
+            echo "No version change, skipping publish"
+            echo "skip=true" >> $GITHUB_OUTPUT
+          else
+            echo "Version changed, proceeding with publish"
+            echo "skip=false" >> $GITHUB_OUTPUT
+          fi
+            
+      - name: Pack package
+        if: steps.check.outputs.skip != 'true'
+        working-directory: packages/client
+        run: npm pack
+        
+      - name: Publish
+        if: steps.check.outputs.skip != 'true'
+        working-directory: packages/client
+        run: npm publish *.tgz --access public

.github/workflows/data-provider.yml

@@ -1,4 +1,4 @@
-name: Node.js Package
+name: Publish `librechat-data-provider` to NPM
 
 on:
   push:
@@ -6,6 +6,12 @@ on:
       - main
     paths:
       - 'packages/data-provider/package.json'
+  workflow_dispatch:
+    inputs:
+      reason:
+        description: 'Reason for manual trigger'
+        required: false
+        default: 'Manual publish requested'
 
 jobs:
   build:
@@ -14,7 +20,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-node@v4
         with:
-          node-version: 16
+          node-version: 20
       - run: cd packages/data-provider && npm ci
       - run: cd packages/data-provider && npm run build
 
@@ -25,7 +31,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-node@v4
         with:
-          node-version: 16
+          node-version: 20
           registry-url: 'https://registry.npmjs.org'
       - run: cd packages/data-provider && npm ci
       - run: cd packages/data-provider && npm run build

.github/workflows/data-schemas.yml

@@ -22,7 +22,7 @@ jobs:
       - name: Use Node.js
         uses: actions/setup-node@v4
         with:
-          node-version: '18.x'
+          node-version: '20.x'
           
       - name: Install dependencies
         run: cd packages/data-schemas && npm ci

.github/workflows/frontend-review.yml

@@ -8,7 +8,7 @@ on:
       - release/*
     paths:
       - 'client/**'
-      - 'packages/**'
+      - 'packages/data-provider/**'
 
 jobs:
   tests_frontend_ubuntu:

.github/workflows/generate-release-changelog-pr.yml

@@ -1,95 +0,0 @@
-name: Generate Release Changelog PR
-
-on:
-  push:
-    tags:
-      - 'v*.*.*'
-  workflow_dispatch:
-
-jobs:
-  generate-release-changelog-pr:
-    permissions:
-      contents: write    # Needed for pushing commits and creating branches.
-      pull-requests: write
-    runs-on: ubuntu-latest
-    steps:
-      # 1. Checkout the repository (with full history).
-      - name: Checkout Repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      # 2. Generate the release changelog using our custom configuration.
-      - name: Generate Release Changelog
-        id: generate_release
-        uses: mikepenz/release-changelog-builder-action@v5.1.0
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        with:
-          configuration: ".github/configuration-release.json"
-          owner: ${{ github.repository_owner }}
-          repo: ${{ github.event.repository.name }}
-          outputFile: CHANGELOG-release.md
-
-      # 3. Update the main CHANGELOG.md:
-      #    - If it doesn't exist, create it with a basic header.
-      #    - Remove the "Unreleased" section (if present).
-      #    - Prepend the new release changelog above previous releases.
-      #    - Remove all temporary files before committing.
-      - name: Update CHANGELOG.md
-        run: |
-          # Determine the release tag, e.g. "v1.2.3"
-          TAG=${GITHUB_REF##*/}
-          echo "Using release tag: $TAG"
-          
-          # Ensure CHANGELOG.md exists; if not, create a basic header.
-          if [ ! -f CHANGELOG.md ]; then
-            echo "# Changelog" > CHANGELOG.md
-            echo "" >> CHANGELOG.md
-            echo "All notable changes to this project will be documented in this file." >> CHANGELOG.md
-            echo "" >> CHANGELOG.md
-          fi
-
-          echo "Updating CHANGELOG.md…"
-          
-          # Remove the "Unreleased" section (from "## [Unreleased]" until the first occurrence of '---') if it exists.
-          if grep -q "^## \[Unreleased\]" CHANGELOG.md; then
-            awk '/^## \[Unreleased\]/{flag=1} flag && /^---/{flag=0; next} !flag' CHANGELOG.md > CHANGELOG.cleaned
-          else
-            cp CHANGELOG.md CHANGELOG.cleaned
-          fi
-
-          # Split the cleaned file into:
-          #   - header.md: content before the first release header ("## [v...").
-          #   - tail.md: content from the first release header onward.
-          awk '/^## \[v/{exit} {print}' CHANGELOG.cleaned > header.md
-          awk 'f{print} /^## \[v/{f=1; print}' CHANGELOG.cleaned > tail.md
-
-          # Combine header, the new release changelog, and the tail.
-          echo "Combining updated changelog parts..."
-          cat header.md CHANGELOG-release.md > CHANGELOG.md.new
-          echo "" >> CHANGELOG.md.new
-          cat tail.md >> CHANGELOG.md.new
-
-          mv CHANGELOG.md.new CHANGELOG.md
-
-          # Remove temporary files.
-          rm -f CHANGELOG.cleaned header.md tail.md CHANGELOG-release.md
-
-          echo "Final CHANGELOG.md content:"
-          cat CHANGELOG.md
-
-      # 4. Create (or update) the Pull Request with the updated CHANGELOG.md.
-      - name: Create Pull Request
-        uses: peter-evans/create-pull-request@v7
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          sign-commits: true
-          commit-message: "chore: update CHANGELOG for release ${{ github.ref_name }}"
-          base: main
-          branch: "changelog/${{ github.ref_name }}"
-          reviewers: danny-avila
-          title: "📜 docs: Changelog for release ${{ github.ref_name }}"
-          body: |
-            **Description**:
-            - This PR updates the CHANGELOG.md by removing the "Unreleased" section and adding new release notes for release ${{ github.ref_name }} above previous releases.

.github/workflows/generate-unreleased-changelog-pr.yml

@@ -1,107 +0,0 @@
-name: Generate Unreleased Changelog PR
-
-on:
-  schedule:
-    - cron: "0 0 * * 1"  # Runs every Monday at 00:00 UTC
-  workflow_dispatch:
-
-jobs:
-  generate-unreleased-changelog-pr:
-    permissions:
-      contents: write    # Needed for pushing commits and creating branches.
-      pull-requests: write
-    runs-on: ubuntu-latest
-    steps:
-      # 1. Checkout the repository on main.
-      - name: Checkout Repository on Main
-        uses: actions/checkout@v4
-        with:
-          ref: main
-          fetch-depth: 0
-
-      # 4. Get the latest version tag.
-      - name: Get Latest Tag
-        id: get_latest_tag
-        run: |
-          LATEST_TAG=$(git describe --tags $(git rev-list --tags --max-count=1) || echo "none")
-          echo "Latest tag: $LATEST_TAG"
-          echo "tag=$LATEST_TAG" >> $GITHUB_OUTPUT
-
-      # 5. Generate the Unreleased changelog.
-      - name: Generate Unreleased Changelog
-        id: generate_unreleased
-        uses: mikepenz/release-changelog-builder-action@v5.1.0
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        with:
-          configuration: ".github/configuration-unreleased.json"
-          owner: ${{ github.repository_owner }}
-          repo: ${{ github.event.repository.name }}
-          outputFile: CHANGELOG-unreleased.md
-          fromTag: ${{ steps.get_latest_tag.outputs.tag }}
-          toTag: main
-
-      # 7. Update CHANGELOG.md with the new Unreleased section.
-      - name: Update CHANGELOG.md
-        id: update_changelog
-        run: |
-          # Create CHANGELOG.md if it doesn't exist.
-          if [ ! -f CHANGELOG.md ]; then
-            echo "# Changelog" > CHANGELOG.md
-            echo "" >> CHANGELOG.md
-            echo "All notable changes to this project will be documented in this file." >> CHANGELOG.md
-            echo "" >> CHANGELOG.md
-          fi
-          
-          echo "Updating CHANGELOG.md…"
-          
-          # Extract content before the "## [Unreleased]" (or first version header if missing).
-          if grep -q "^## \[Unreleased\]" CHANGELOG.md; then
-            awk '/^## \[Unreleased\]/{exit} {print}' CHANGELOG.md > CHANGELOG_TMP.md
-          else
-            awk '/^## \[v/{exit} {print}' CHANGELOG.md > CHANGELOG_TMP.md
-          fi
-          
-          # Append the generated Unreleased changelog.
-          echo "" >> CHANGELOG_TMP.md
-          cat CHANGELOG-unreleased.md >> CHANGELOG_TMP.md
-          echo "" >> CHANGELOG_TMP.md
-          
-          # Append the remainder of the original changelog (starting from the first version header).
-          awk 'f{print} /^## \[v/{f=1; print}' CHANGELOG.md >> CHANGELOG_TMP.md
-          
-          # Replace the old file with the updated file.
-          mv CHANGELOG_TMP.md CHANGELOG.md
-          
-          # Remove the temporary generated file.
-          rm -f CHANGELOG-unreleased.md
-          
-          echo "Final CHANGELOG.md:"
-          cat CHANGELOG.md
-
-      # 8. Check if CHANGELOG.md has any updates.
-      - name: Check for CHANGELOG.md changes
-        id: changelog_changes
-        run: |
-          if git diff --quiet CHANGELOG.md; then
-            echo "has_changes=false" >> $GITHUB_OUTPUT
-          else
-            echo "has_changes=true" >> $GITHUB_OUTPUT
-          fi
-
-      # 9. Create (or update) the Pull Request only if there are changes.
-      - name: Create Pull Request
-        if: steps.changelog_changes.outputs.has_changes == 'true'
-        uses: peter-evans/create-pull-request@v7
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          base: main
-          branch: "changelog/unreleased-update"
-          sign-commits: true
-          commit-message: "action: update Unreleased changelog"
-          title: "📜 docs: Unreleased Changelog"
-          body: |
-            **Description**:
-            - This PR updates the Unreleased section in CHANGELOG.md.
-            - It compares the current main branch with the latest version tag (determined as ${{ steps.get_latest_tag.outputs.tag }}),
-              regenerates the Unreleased changelog, removes any old Unreleased block, and inserts the new content.

.github/workflows/helmcharts.yml

@@ -4,12 +4,13 @@ name: Build Helm Charts on Tag
 on:
   push:
     tags:
-      - "*"
+      - "chart-*"
 
 jobs:
   release:
     permissions:
       contents: write
+      packages: write
     runs-on: ubuntu-latest
     steps:
       - name: Checkout
@@ -26,15 +27,49 @@ jobs:
         uses: azure/setup-helm@v4
         env:
           GITHUB_TOKEN: "${{ secrets.GITHUB_TOKEN }}"
+      
       - name: Build Subchart Deps
         run: |
-          cd helm/librechat-rag-api
-          helm dependency build 
+          cd helm/librechat
+          helm dependency build  
+          cd ../librechat-rag-api
+          helm dependency build      
 
-      - name: Run chart-releaser
-        uses: helm/chart-releaser-action@v1.6.0
+      - name: Get Chart Version
+        id: chart-version
+        run: |
+          CHART_VERSION=$(echo "${{ github.ref_name }}" | cut -d'-' -f2)
+          echo "CHART_VERSION=${CHART_VERSION}" >> "$GITHUB_OUTPUT"
+
+      # Log in to GitHub Container Registry
+      - name: Log in to GitHub Container Registry
+        uses: docker/login-action@v3
         with:
-          charts_dir: helm
-          skip_existing: true
-        env:
-          CR_TOKEN: "${{ secrets.GITHUB_TOKEN }}"
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      # Run Helm OCI Charts Releaser
+      # This is for the librechat chart
+      - name: Release Helm OCI Charts for librechat
+        uses: appany/helm-oci-chart-releaser@v0.4.2
+        with:
+          name: librechat
+          repository: ${{ github.actor }}/librechat-chart
+          tag: ${{ steps.chart-version.outputs.CHART_VERSION }}
+          path: helm/librechat
+          registry: ghcr.io
+          registry_username: ${{ github.actor }}
+          registry_password: ${{ secrets.GITHUB_TOKEN }}
+      
+      # this is for the librechat-rag-api chart
+      - name: Release Helm OCI Charts for librechat-rag-api
+        uses: appany/helm-oci-chart-releaser@v0.4.2
+        with:
+          name: librechat-rag-api
+          repository: ${{ github.actor }}/librechat-chart
+          tag: ${{ steps.chart-version.outputs.CHART_VERSION }}
+          path: helm/librechat-rag-api
+          registry: ghcr.io
+          registry_username: ${{ github.actor }}
+          registry_password: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/i18n-unused-keys.yml

@@ -1,11 +1,18 @@
 name: Detect Unused i18next Strings
 
+# This workflow checks for unused i18n keys in translation files.
+# It has special handling for:
+# - com_ui_special_var_* keys that are dynamically constructed
+# - com_agents_category_* keys that are stored in the database and used dynamically
+
 on:
   pull_request:
     paths:
       - "client/src/**"
       - "api/**"
       - "packages/data-provider/src/**"
+      - "packages/client/**"
+      - "packages/data-schemas/src/**"
 
 jobs:
   detect-unused-i18n-keys:
@@ -23,7 +30,7 @@ jobs:
 
           # Define paths
           I18N_FILE="client/src/locales/en/translation.json"
-          SOURCE_DIRS=("client/src" "api" "packages/data-provider/src")
+          SOURCE_DIRS=("client/src" "api" "packages/data-provider/src" "packages/client" "packages/data-schemas/src")
 
           # Check if translation file exists
           if [[ ! -f "$I18N_FILE" ]]; then
@@ -51,6 +58,31 @@ jobs:
                 fi
               done
               
+              # Also check if the key is directly used somewhere
+              if [[ "$FOUND" == false ]]; then
+                for DIR in "${SOURCE_DIRS[@]}"; do
+                  if grep -r --include=\*.{js,jsx,ts,tsx} -q "$KEY" "$DIR"; then
+                    FOUND=true
+                    break
+                  fi
+                done
+              fi
+            # Special case for agent category keys that are dynamically used from database
+            elif [[ "$KEY" == com_agents_category_* ]]; then
+              # Check if agent category localization is being used
+              for DIR in "${SOURCE_DIRS[@]}"; do
+                # Check for dynamic category label/description usage
+                if grep -r --include=\*.{js,jsx,ts,tsx} -E "category\.(label|description).*startsWith.*['\"]com_" "$DIR" > /dev/null 2>&1 || \
+                   # Check for the method that defines these keys
+                   grep -r --include=\*.{js,jsx,ts,tsx} "ensureDefaultCategories" "$DIR" > /dev/null 2>&1 || \
+                   # Check for direct usage in agentCategory.ts
+                   grep -r --include=\*.ts -E "label:.*['\"]$KEY['\"]" "$DIR" > /dev/null 2>&1 || \
+                   grep -r --include=\*.ts -E "description:.*['\"]$KEY['\"]" "$DIR" > /dev/null 2>&1; then
+                  FOUND=true
+                  break
+                fi
+              done
+              
               # Also check if the key is directly used somewhere
               if [[ "$FOUND" == false ]]; then
                 for DIR in "${SOURCE_DIRS[@]}"; do

.github/workflows/locize-i18n-sync.yml

@@ -48,7 +48,7 @@ jobs:
 
       # 2. Download translation files from locize.
       - name: Download Translations from locize
-        uses: locize/download@v1
+        uses: locize/download@v2
         with:
           project-id: ${{ secrets.LOCIZE_PROJECT_ID }}
           path: "client/src/locales"

.github/workflows/unused-packages.yml

@@ -7,6 +7,7 @@ on:
       - 'package-lock.json'
       - 'client/**'
       - 'api/**'
+      - 'packages/client/**'
 
 jobs:
   detect-unused-packages:
@@ -28,7 +29,7 @@ jobs:
 
       - name: Validate JSON files
         run: |
-          for FILE in package.json client/package.json api/package.json; do
+          for FILE in package.json client/package.json api/package.json packages/client/package.json; do
             if [[ -f "$FILE" ]]; then
               jq empty "$FILE" || (echo "::error title=Invalid JSON::$FILE is invalid" && exit 1)
             fi
@@ -63,12 +64,31 @@ jobs:
             local folder=$1
             local output_file=$2
             if [[ -d "$folder" ]]; then
-              grep -rEho "require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)" "$folder" --include=\*.{js,ts,mjs,cjs} | \
+              # Extract require() statements
+              grep -rEho "require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
                 sed -E "s/require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)/\1/" > "$output_file"
           
-              grep -rEho "import .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,mjs,cjs} | \
+              # Extract ES6 imports - various patterns
+              # import x from 'module'
+              grep -rEho "import .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
                 sed -E "s/import .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]/\1/" >> "$output_file"
+              
+              # import 'module' (side-effect imports)
+              grep -rEho "import ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
+                sed -E "s/import ['\"]([a-zA-Z0-9@/._-]+)['\"]/\1/" >> "$output_file"
+              
+              # export { x } from 'module' or export * from 'module'
+              grep -rEho "export .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
+                sed -E "s/export .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]/\1/" >> "$output_file"
+              
+              # import type { x } from 'module' (TypeScript)
+              grep -rEho "import type .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{ts,tsx} | \
+                sed -E "s/import type .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]/\1/" >> "$output_file"
           
+              # Remove subpath imports but keep the base package
+              # e.g., '@tanstack/react-query/devtools' becomes '@tanstack/react-query'
+              sed -i -E 's|^(@?[a-zA-Z0-9-]+(/[a-zA-Z0-9-]+)?)/.*|\1|' "$output_file"
+              
               sort -u "$output_file" -o "$output_file"
             else
               touch "$output_file"
@@ -78,13 +98,80 @@ jobs:
           extract_deps_from_code "." root_used_code.txt
           extract_deps_from_code "client" client_used_code.txt
           extract_deps_from_code "api" api_used_code.txt
+          
+          # Extract dependencies used by @librechat/client package
+          extract_deps_from_code "packages/client" packages_client_used_code.txt
+
+      - name: Get @librechat/client dependencies
+        id: get-librechat-client-deps
+        run: |
+          if [[ -f "packages/client/package.json" ]]; then
+            # Get all dependencies from @librechat/client (dependencies, devDependencies, and peerDependencies)
+            DEPS=$(jq -r '.dependencies // {} | keys[]' packages/client/package.json 2>/dev/null || echo "")
+            DEV_DEPS=$(jq -r '.devDependencies // {} | keys[]' packages/client/package.json 2>/dev/null || echo "")
+            PEER_DEPS=$(jq -r '.peerDependencies // {} | keys[]' packages/client/package.json 2>/dev/null || echo "")
+            
+            # Combine all dependencies
+            echo "$DEPS" > librechat_client_deps.txt
+            echo "$DEV_DEPS" >> librechat_client_deps.txt
+            echo "$PEER_DEPS" >> librechat_client_deps.txt
+            
+            # Also include dependencies that are imported in packages/client
+            cat packages_client_used_code.txt >> librechat_client_deps.txt
+            
+            # Remove empty lines and sort
+            grep -v '^$' librechat_client_deps.txt | sort -u > temp_deps.txt
+            mv temp_deps.txt librechat_client_deps.txt
+          else
+            touch librechat_client_deps.txt
+          fi
+
+      - name: Extract Workspace Dependencies
+        id: extract-workspace-deps
+        run: |
+          # Function to get dependencies from a workspace package that are used by another package
+          get_workspace_package_deps() {
+            local package_json=$1
+            local output_file=$2
+            
+            # Get all workspace dependencies (starting with @librechat/)
+            if [[ -f "$package_json" ]]; then
+              local workspace_deps=$(jq -r '.dependencies // {} | to_entries[] | select(.key | startswith("@librechat/")) | .key' "$package_json" 2>/dev/null || echo "")
+              
+              # For each workspace dependency, get its dependencies
+              for dep in $workspace_deps; do
+                # Convert @librechat/api to packages/api
+                local workspace_path=$(echo "$dep" | sed 's/@librechat\//packages\//')
+                local workspace_package_json="${workspace_path}/package.json"
+                
+                if [[ -f "$workspace_package_json" ]]; then
+                  # Extract all dependencies from the workspace package
+                  jq -r '.dependencies // {} | keys[]' "$workspace_package_json" 2>/dev/null >> "$output_file"
+                  # Also extract peerDependencies
+                  jq -r '.peerDependencies // {} | keys[]' "$workspace_package_json" 2>/dev/null >> "$output_file"
+                fi
+              done
+            fi
+            
+            if [[ -f "$output_file" ]]; then
+              sort -u "$output_file" -o "$output_file"
+            else
+              touch "$output_file"
+            fi
+          }
+          
+          # Get workspace dependencies for each package
+          get_workspace_package_deps "package.json" root_workspace_deps.txt
+          get_workspace_package_deps "client/package.json" client_workspace_deps.txt
+          get_workspace_package_deps "api/package.json" api_workspace_deps.txt
 
       - name: Run depcheck for root package.json
         id: check-root
         run: |
           if [[ -f "package.json" ]]; then
             UNUSED=$(depcheck --json | jq -r '.dependencies | join("\n")' || echo "")
-            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat root_used_deps.txt root_used_code.txt | sort) || echo "")
+            # Exclude dependencies used in scripts, code, and workspace packages
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat root_used_deps.txt root_used_code.txt root_workspace_deps.txt | sort) || echo "")
             echo "ROOT_UNUSED<<EOF" >> $GITHUB_ENV
             echo "$UNUSED" >> $GITHUB_ENV
             echo "EOF" >> $GITHUB_ENV
@@ -97,7 +184,10 @@ jobs:
             chmod -R 755 client
             cd client
             UNUSED=$(depcheck --json | jq -r '.dependencies | join("\n")' || echo "")
-            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../client_used_deps.txt ../client_used_code.txt | sort) || echo "")
+            # Exclude dependencies used in scripts, code, and workspace packages
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../client_used_deps.txt ../client_used_code.txt ../client_workspace_deps.txt | sort) || echo "")
+            # Filter out false positives
+            UNUSED=$(echo "$UNUSED" | grep -v "^micromark-extension-llm-math$" || echo "")
             echo "CLIENT_UNUSED<<EOF" >> $GITHUB_ENV
             echo "$UNUSED" >> $GITHUB_ENV
             echo "EOF" >> $GITHUB_ENV
@@ -111,7 +201,8 @@ jobs:
             chmod -R 755 api
             cd api
             UNUSED=$(depcheck --json | jq -r '.dependencies | join("\n")' || echo "")
-            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../api_used_deps.txt ../api_used_code.txt | sort) || echo "")
+            # Exclude dependencies used in scripts, code, and workspace packages
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../api_used_deps.txt ../api_used_code.txt ../api_workspace_deps.txt | sort) || echo "")
             echo "API_UNUSED<<EOF" >> $GITHUB_ENV
             echo "$UNUSED" >> $GITHUB_ENV
             echo "EOF" >> $GITHUB_ENV

.gitignore

@@ -13,6 +13,9 @@ pids
 *.seed
 .git
 
+# CI/CD data
+test-image*
+
 # Directory for instrumented libs generated by jscoverage/JSCover
 lib-cov
 
@@ -55,6 +58,7 @@ bower_components/
 # AI
 .clineignore
 .cursor
+.aider*
 
 # Floobits
 .floo
@@ -124,3 +128,13 @@ helm/**/.values.yaml
 
 # SAML Idp cert
 *.cert
+
+# AI Assistants
+/.claude/
+/.cursor/
+/.copilot/
+/.aider/
+/.openai/
+/.tabnine/
+/.codeium
+*.local.md

.vscode/launch.json

@@ -8,7 +8,8 @@
       "skipFiles": ["<node_internals>/**"],
       "program": "${workspaceFolder}/api/server/index.js",
       "env": {
-        "NODE_ENV": "production"
+        "NODE_ENV": "production",
+        "NODE_TLS_REJECT_UNAUTHORIZED": "0"
       },
       "console": "integratedTerminal",
       "envFile": "${workspaceFolder}/.env"

Dockerfile

@@ -1,4 +1,4 @@
-# v0.7.8
+# v0.8.0-rc4
 
 # Base node image
 FROM node:20-alpine AS node
@@ -19,24 +19,31 @@ WORKDIR /app
 
 USER node
 
-COPY --chown=node:node . .
+COPY --chown=node:node package.json package-lock.json ./
+COPY --chown=node:node api/package.json ./api/package.json
+COPY --chown=node:node client/package.json ./client/package.json
+COPY --chown=node:node packages/data-provider/package.json ./packages/data-provider/package.json
+COPY --chown=node:node packages/data-schemas/package.json ./packages/data-schemas/package.json
+COPY --chown=node:node packages/api/package.json ./packages/api/package.json
 
 RUN \
     # Allow mounting of these files, which have no default
     touch .env ; \
     # Create directories for the volumes to inherit the correct permissions
-    mkdir -p /app/client/public/images /app/api/logs ; \
+    mkdir -p /app/client/public/images /app/api/logs /app/uploads ; \
     npm config set fetch-retry-maxtimeout 600000 ; \
     npm config set fetch-retries 5 ; \
     npm config set fetch-retry-mintimeout 15000 ; \
-    npm install --no-audit; \
+    npm ci --no-audit
+
+COPY --chown=node:node . .
+
+RUN \
     # React client build
     NODE_OPTIONS="--max-old-space-size=2048" npm run frontend; \
     npm prune --production; \
     npm cache clean --force
 
-RUN mkdir -p /app/client/public/images /app/api/logs
-
 # Node API setup
 EXPOSE 3080
 ENV HOST=0.0.0.0
@@ -47,4 +54,4 @@ CMD ["npm", "run", "backend"]
 # WORKDIR /usr/share/nginx/html
 # COPY --from=node /app/client/dist /usr/share/nginx/html
 # COPY client/nginx.conf /etc/nginx/conf.d/default.conf
-# ENTRYPOINT ["nginx", "-g", "daemon off;"]
+# ENTRYPOINT ["nginx", "-g", "daemon off;"]

Dockerfile.multi

@@ -1,5 +1,5 @@
 # Dockerfile.multi
-# v0.7.8
+# v0.8.0-rc4
 
 # Base for all builds
 FROM node:20-alpine AS base-min
@@ -16,6 +16,7 @@ COPY package*.json ./
 COPY packages/data-provider/package*.json ./packages/data-provider/
 COPY packages/api/package*.json ./packages/api/
 COPY packages/data-schemas/package*.json ./packages/data-schemas/
+COPY packages/client/package*.json ./packages/client/
 COPY client/package*.json ./client/
 COPY api/package*.json ./api/
 
@@ -45,11 +46,19 @@ COPY --from=data-provider-build /app/packages/data-provider/dist /app/packages/d
 COPY --from=data-schemas-build /app/packages/data-schemas/dist /app/packages/data-schemas/dist
 RUN npm run build
 
+# Build `client` package
+FROM base AS client-package-build
+WORKDIR /app/packages/client
+COPY packages/client ./
+RUN npm run build
+
 # Client build
 FROM base AS client-build
 WORKDIR /app/client
 COPY client ./
 COPY --from=data-provider-build /app/packages/data-provider/dist /app/packages/data-provider/dist
+COPY --from=client-package-build /app/packages/client/dist /app/packages/client/dist
+COPY --from=client-package-build /app/packages/client/src /app/packages/client/src
 ENV NODE_OPTIONS="--max-old-space-size=2048"
 RUN npm run build
 

README.md

@@ -52,7 +52,7 @@
 - 🖥️ **UI & Experience** inspired by ChatGPT with enhanced design and features
 
 - 🤖 **AI Model Selection**:  
-  - Anthropic (Claude), AWS Bedrock, OpenAI, Azure OpenAI, Google, Vertex AI, OpenAI Assistants API (incl. Azure)
+  - Anthropic (Claude), AWS Bedrock, OpenAI, Azure OpenAI, Google, Vertex AI, OpenAI Responses API (incl. Azure)
   - [Custom Endpoints](https://www.librechat.ai/docs/quick_start/custom_endpoints): Use any OpenAI-compatible API with LibreChat, no proxy required
   - Compatible with [Local & Remote AI Providers](https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints):
     - Ollama, groq, Cohere, Mistral AI, Apple MLX, koboldcpp, together.ai,
@@ -65,15 +65,17 @@
 
 - 🔦 **Agents & Tools Integration**:  
   - **[LibreChat Agents](https://www.librechat.ai/docs/features/agents)**:
-    - No-Code Custom Assistants: Build specialized, AI-driven helpers without coding  
-    - Flexible & Extensible: Attach tools like DALL-E-3, file search, code execution, and more  
-    - Compatible with Custom Endpoints, OpenAI, Azure, Anthropic, AWS Bedrock, and more
+    - No-Code Custom Assistants: Build specialized, AI-driven helpers
+    - Agent Marketplace: Discover and deploy community-built agents
+    - Collaborative Sharing: Share agents with specific users and groups
+    - Flexible & Extensible: Use MCP Servers, tools, file search, code execution, and more
+    - Compatible with Custom Endpoints, OpenAI, Azure, Anthropic, AWS Bedrock, Google, Vertex AI, Responses API, and more
     - [Model Context Protocol (MCP) Support](https://modelcontextprotocol.io/clients#librechat) for Tools
-  - Use LibreChat Agents and OpenAI Assistants with Files, Code Interpreter, Tools, and API Actions
 
 - 🔍 **Web Search**:  
   - Search the internet and retrieve relevant information to enhance your AI context
   - Combines search providers, content scrapers, and result rerankers for optimal results
+  - **Customizable Jina Reranking**: Configure custom Jina API URLs for reranking services
   - **[Learn More →](https://www.librechat.ai/docs/features/web_search)**
 
 - 🪄 **Generative UI with Code Artifacts**:  
@@ -88,15 +90,18 @@
   - Create, Save, & Share Custom Presets  
   - Switch between AI Endpoints and Presets mid-chat
   - Edit, Resubmit, and Continue Messages with Conversation branching  
+  - Create and share prompts with specific users and groups
   - [Fork Messages & Conversations](https://www.librechat.ai/docs/features/fork) for Advanced Context control
 
 - 💬 **Multimodal & File Interactions**:  
   - Upload and analyze images with Claude 3, GPT-4.5, GPT-4o, o1, Llama-Vision, and Gemini 📸  
   - Chat with Files using Custom Endpoints, OpenAI, Azure, Anthropic, AWS Bedrock, & Google 🗃️
 
-- 🌎 **Multilingual UI**:  
-  - English, 中文, Deutsch, Español, Français, Italiano, Polski, Português Brasileiro
-  - Русский, 日本語, Svenska, 한국어, Tiếng Việt, 繁體中文, العربية, Türkçe, Nederlands, עברית
+- 🌎 **Multilingual UI**:
+  - English, 中文 (简体), 中文 (繁體), العربية, Deutsch, Español, Français, Italiano
+  - Polski, Português (PT), Português (BR), Русский, 日本語, Svenska, 한국어, Tiếng Việt
+  - Türkçe, Nederlands, עברית, Català, Čeština, Dansk, Eesti, فارسی
+  - Suomi, Magyar, Հայերեն, Bahasa Indonesia, ქართული, Latviešu, ไทย, ئۇيغۇرچە
 
 - 🧠 **Reasoning UI**:  
   - Dynamic Reasoning UI for Chain-of-Thought/Reasoning AI models like DeepSeek-R1

api/app/clients/AnthropicClient.js

@@ -10,7 +10,17 @@ const {
   validateVisionModel,
 } = require('librechat-data-provider');
 const { SplitStreamHandler: _Handler } = require('@librechat/agents');
-const { Tokenizer, createFetch, createStreamEventHandlers } = require('@librechat/api');
+const {
+  Tokenizer,
+  createFetch,
+  matchModelName,
+  getClaudeHeaders,
+  getModelMaxTokens,
+  configureReasoning,
+  checkPromptCacheSupport,
+  getModelMaxOutputTokens,
+  createStreamEventHandlers,
+} = require('@librechat/api');
 const {
   truncateText,
   formatMessage,
@@ -19,12 +29,6 @@ const {
   parseParamFromPrompt,
   createContextHandlers,
 } = require('./prompts');
-const {
-  getClaudeHeaders,
-  configureReasoning,
-  checkPromptCacheSupport,
-} = require('~/server/services/Endpoints/anthropic/helpers');
-const { getModelMaxTokens, getModelMaxOutputTokens, matchModelName } = require('~/utils');
 const { spendTokens, spendStructuredTokens } = require('~/models/spendTokens');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
 const { sleep } = require('~/server/utils');
@@ -190,10 +194,11 @@ class AnthropicClient extends BaseClient {
         reverseProxyUrl: this.options.reverseProxyUrl,
       }),
       apiKey: this.apiKey,
+      fetchOptions: {},
     };
 
     if (this.options.proxy) {
-      options.httpAgent = new HttpsProxyAgent(this.options.proxy);
+      options.fetchOptions.agent = new HttpsProxyAgent(this.options.proxy);
     }
 
     if (this.options.reverseProxyUrl) {

api/app/clients/BaseClient.js

@@ -1,5 +1,7 @@
 const crypto = require('crypto');
 const fetch = require('node-fetch');
+const { logger } = require('@librechat/data-schemas');
+const { getBalanceConfig } = require('@librechat/api');
 const {
   supportsBalanceCheck,
   isAgentsEndpoint,
@@ -13,10 +15,8 @@ const {
 const { getMessages, saveMessage, updateMessage, saveConvo, getConvo } = require('~/models');
 const { checkBalance } = require('~/models/balanceMethods');
 const { truncateToolCallOutputs } = require('./prompts');
-const { addSpaceIfNeeded } = require('~/server/utils');
 const { getFiles } = require('~/models/File');
 const TextStream = require('./TextStream');
-const { logger } = require('~/config');
 
 class BaseClient {
   constructor(apiKey, options = {}) {
@@ -38,6 +38,8 @@ class BaseClient {
     this.conversationId;
     /** @type {string} */
     this.responseMessageId;
+    /** @type {string} */
+    this.parentMessageId;
     /** @type {TAttachment[]} */
     this.attachments;
     /** The key for the usage object's input tokens
@@ -109,12 +111,17 @@ class BaseClient {
   /**
    * Abstract method to record token usage. Subclasses must implement this method.
    * If a correction to the token usage is needed, the method should return an object with the corrected token counts.
+   * Should only be used if `recordCollectedUsage` was not used instead.
+   * @param {string} [model]
+   * @param {AppConfig['balance']} [balance]
    * @param {number} promptTokens
    * @param {number} completionTokens
    * @returns {Promise<void>}
    */
-  async recordTokenUsage({ promptTokens, completionTokens }) {
+  async recordTokenUsage({ model, balance, promptTokens, completionTokens }) {
     logger.debug('[BaseClient] `recordTokenUsage` not implemented.', {
+      model,
+      balance,
       promptTokens,
       completionTokens,
     });
@@ -183,7 +190,8 @@ class BaseClient {
     this.user = user;
     const saveOptions = this.getSaveOptions();
     this.abortController = opts.abortController ?? new AbortController();
-    const conversationId = overrideConvoId ?? opts.conversationId ?? crypto.randomUUID();
+    const requestConvoId = overrideConvoId ?? opts.conversationId;
+    const conversationId = requestConvoId ?? crypto.randomUUID();
     const parentMessageId = opts.parentMessageId ?? Constants.NO_PARENT;
     const userMessageId =
       overrideUserMessageId ?? opts.overrideParentMessageId ?? crypto.randomUUID();
@@ -198,17 +206,22 @@ class BaseClient {
       this.currentMessages[this.currentMessages.length - 1].messageId = head;
     }
 
+    if (opts.isRegenerate && responseMessageId.endsWith('_')) {
+      responseMessageId = crypto.randomUUID();
+    }
+
     this.responseMessageId = responseMessageId;
 
     return {
       ...opts,
       user,
       head,
+      saveOptions,
+      userMessageId,
+      requestConvoId,
       conversationId,
       parentMessageId,
-      userMessageId,
       responseMessageId,
-      saveOptions,
     };
   }
 
@@ -227,11 +240,12 @@ class BaseClient {
     const {
       user,
       head,
+      saveOptions,
+      userMessageId,
+      requestConvoId,
       conversationId,
       parentMessageId,
-      userMessageId,
       responseMessageId,
-      saveOptions,
     } = await this.setMessageOptions(opts);
 
     const userMessage = opts.isEdited
@@ -253,7 +267,8 @@ class BaseClient {
     }
 
     if (typeof opts?.onStart === 'function') {
-      opts.onStart(userMessage, responseMessageId);
+      const isNewConvo = !requestConvoId && parentMessageId === Constants.NO_PARENT;
+      opts.onStart(userMessage, responseMessageId, isNewConvo);
     }
 
     return {
@@ -559,6 +574,7 @@ class BaseClient {
   }
 
   async sendMessage(message, opts = {}) {
+    const appConfig = this.options.req?.config;
     /** @type {Promise<TMessage>} */
     let userMessagePromise;
     const { user, head, isEdited, conversationId, responseMessageId, saveOptions, userMessage } =
@@ -572,7 +588,7 @@ class BaseClient {
       });
     }
 
-    const { generation = '' } = opts;
+    const { editedContent } = opts;
 
     // It's not necessary to push to currentMessages
     // depending on subclass implementation of handling messages
@@ -587,26 +603,40 @@ class BaseClient {
           isCreatedByUser: false,
           model: this.modelOptions?.model ?? this.model,
           sender: this.sender,
-          text: generation,
         };
         this.currentMessages.push(userMessage, latestMessage);
-      } else {
-        latestMessage.text = generation;
+      } else if (editedContent != null) {
+        // Handle editedContent for content parts
+        if (editedContent && latestMessage.content && Array.isArray(latestMessage.content)) {
+          const { index, text, type } = editedContent;
+          if (index >= 0 && index < latestMessage.content.length) {
+            const contentPart = latestMessage.content[index];
+            if (type === ContentTypes.THINK && contentPart.type === ContentTypes.THINK) {
+              contentPart[ContentTypes.THINK] = text;
+            } else if (type === ContentTypes.TEXT && contentPart.type === ContentTypes.TEXT) {
+              contentPart[ContentTypes.TEXT] = text;
+            }
+          }
+        }
       }
       this.continued = true;
     } else {
       this.currentMessages.push(userMessage);
     }
 
+    /**
+     * When the userMessage is pushed to currentMessages, the parentMessage is the userMessageId.
+     * this only matters when buildMessages is utilizing the parentMessageId, and may vary on implementation
+     */
+    const parentMessageId = isEdited ? head : userMessage.messageId;
+    this.parentMessageId = parentMessageId;
     let {
       prompt: payload,
       tokenCountMap,
       promptTokens,
     } = await this.buildMessages(
       this.currentMessages,
-      // When the userMessage is pushed to currentMessages, the parentMessage is the userMessageId.
-      // this only matters when buildMessages is utilizing the parentMessageId, and may vary on implementation
-      isEdited ? head : userMessage.messageId,
+      parentMessageId,
       this.getBuildMessagesOptions(opts),
       opts,
     );
@@ -631,9 +661,9 @@ class BaseClient {
       }
     }
 
-    const balance = this.options.req?.app?.locals?.balance;
+    const balanceConfig = getBalanceConfig(appConfig);
     if (
-      balance?.enabled &&
+      balanceConfig?.enabled &&
       supportsBalanceCheck[this.options.endpointType ?? this.options.endpoint]
     ) {
       await checkBalance({
@@ -672,16 +702,32 @@ class BaseClient {
     };
 
     if (typeof completion === 'string') {
-      responseMessage.text = addSpaceIfNeeded(generation) + completion;
+      responseMessage.text = completion;
     } else if (
       Array.isArray(completion) &&
       (this.clientName === EModelEndpoint.agents ||
         isParamEndpoint(this.options.endpoint, this.options.endpointType))
     ) {
       responseMessage.text = '';
-      responseMessage.content = completion;
+
+      if (!opts.editedContent || this.currentMessages.length === 0) {
+        responseMessage.content = completion;
+      } else {
+        const latestMessage = this.currentMessages[this.currentMessages.length - 1];
+        if (!latestMessage?.content) {
+          responseMessage.content = completion;
+        } else {
+          const existingContent = [...latestMessage.content];
+          const { type: editedType } = opts.editedContent;
+          responseMessage.content = this.mergeEditedContent(
+            existingContent,
+            completion,
+            editedType,
+          );
+        }
+      }
     } else if (Array.isArray(completion)) {
-      responseMessage.text = addSpaceIfNeeded(generation) + completion.join('');
+      responseMessage.text = completion.join('');
     }
 
     if (
@@ -712,9 +758,14 @@ class BaseClient {
       } else {
         responseMessage.tokenCount = this.getTokenCountForResponse(responseMessage);
         completionTokens = responseMessage.tokenCount;
+        await this.recordTokenUsage({
+          usage,
+          promptTokens,
+          completionTokens,
+          balance: balanceConfig,
+          model: responseMessage.model,
+        });
       }
-
-      await this.recordTokenUsage({ promptTokens, completionTokens, usage });
     }
 
     if (userMessagePromise) {
@@ -792,7 +843,8 @@ class BaseClient {
 
     userMessage.tokenCount = userMessageTokenCount;
     /*
-      Note: `AskController` saves the user message, so we update the count of its `userMessage` reference
+      Note: `AgentController` saves the user message if not saved here
+      (noted by `savedMessageIds`), so we update the count of its `userMessage` reference
     */
     if (typeof opts?.getReqData === 'function') {
       opts.getReqData({
@@ -801,7 +853,8 @@ class BaseClient {
     }
     /*
       Note: we update the user message to be sure it gets the calculated token count;
-      though `AskController` saves the user message, EditController does not
+      though `AgentController` saves the user message if not saved here
+      (noted by `savedMessageIds`), EditController does not
     */
     await userMessagePromise;
     await this.updateMessageInDatabase({
@@ -1093,6 +1146,50 @@ class BaseClient {
     return numTokens;
   }
 
+  /**
+   * Merges completion content with existing content when editing TEXT or THINK types
+   * @param {Array} existingContent - The existing content array
+   * @param {Array} newCompletion - The new completion content
+   * @param {string} editedType - The type of content being edited
+   * @returns {Array} The merged content array
+   */
+  mergeEditedContent(existingContent, newCompletion, editedType) {
+    if (!newCompletion.length) {
+      return existingContent.concat(newCompletion);
+    }
+
+    if (editedType !== ContentTypes.TEXT && editedType !== ContentTypes.THINK) {
+      return existingContent.concat(newCompletion);
+    }
+
+    const lastIndex = existingContent.length - 1;
+    const lastExisting = existingContent[lastIndex];
+    const firstNew = newCompletion[0];
+
+    if (lastExisting?.type !== firstNew?.type || firstNew?.type !== editedType) {
+      return existingContent.concat(newCompletion);
+    }
+
+    const mergedContent = [...existingContent];
+    if (editedType === ContentTypes.TEXT) {
+      mergedContent[lastIndex] = {
+        ...mergedContent[lastIndex],
+        [ContentTypes.TEXT]:
+          (mergedContent[lastIndex][ContentTypes.TEXT] || '') + (firstNew[ContentTypes.TEXT] || ''),
+      };
+    } else {
+      mergedContent[lastIndex] = {
+        ...mergedContent[lastIndex],
+        [ContentTypes.THINK]:
+          (mergedContent[lastIndex][ContentTypes.THINK] || '') +
+          (firstNew[ContentTypes.THINK] || ''),
+      };
+    }
+
+    // Add remaining completion items
+    return mergedContent.concat(newCompletion.slice(1));
+  }
+
   async sendPayload(payload, opts = {}) {
     if (opts && typeof opts === 'object') {
       this.setOptions(opts);

api/app/clients/ChatGPTClient.js

@@ -1,804 +0,0 @@
-const { Keyv } = require('keyv');
-const crypto = require('crypto');
-const { CohereClient } = require('cohere-ai');
-const { fetchEventSource } = require('@waylaidwanderer/fetch-event-source');
-const { constructAzureURL, genAzureChatCompletion } = require('@librechat/api');
-const { encoding_for_model: encodingForModel, get_encoding: getEncoding } = require('tiktoken');
-const {
-  ImageDetail,
-  EModelEndpoint,
-  resolveHeaders,
-  CohereConstants,
-  mapModelToAzureConfig,
-} = require('librechat-data-provider');
-const { createContextHandlers } = require('./prompts');
-const { createCoherePayload } = require('./llm');
-const { extractBaseURL } = require('~/utils');
-const BaseClient = require('./BaseClient');
-const { logger } = require('~/config');
-
-const CHATGPT_MODEL = 'gpt-3.5-turbo';
-const tokenizersCache = {};
-
-class ChatGPTClient extends BaseClient {
-  constructor(apiKey, options = {}, cacheOptions = {}) {
-    super(apiKey, options, cacheOptions);
-
-    cacheOptions.namespace = cacheOptions.namespace || 'chatgpt';
-    this.conversationsCache = new Keyv(cacheOptions);
-    this.setOptions(options);
-  }
-
-  setOptions(options) {
-    if (this.options && !this.options.replaceOptions) {
-      // nested options aren't spread properly, so we need to do this manually
-      this.options.modelOptions = {
-        ...this.options.modelOptions,
-        ...options.modelOptions,
-      };
-      delete options.modelOptions;
-      // now we can merge options
-      this.options = {
-        ...this.options,
-        ...options,
-      };
-    } else {
-      this.options = options;
-    }
-
-    if (this.options.openaiApiKey) {
-      this.apiKey = this.options.openaiApiKey;
-    }
-
-    const modelOptions = this.options.modelOptions || {};
-    this.modelOptions = {
-      ...modelOptions,
-      // set some good defaults (check for undefined in some cases because they may be 0)
-      model: modelOptions.model || CHATGPT_MODEL,
-      temperature: typeof modelOptions.temperature === 'undefined' ? 0.8 : modelOptions.temperature,
-      top_p: typeof modelOptions.top_p === 'undefined' ? 1 : modelOptions.top_p,
-      presence_penalty:
-        typeof modelOptions.presence_penalty === 'undefined' ? 1 : modelOptions.presence_penalty,
-      stop: modelOptions.stop,
-    };
-
-    this.isChatGptModel = this.modelOptions.model.includes('gpt-');
-    const { isChatGptModel } = this;
-    this.isUnofficialChatGptModel =
-      this.modelOptions.model.startsWith('text-chat') ||
-      this.modelOptions.model.startsWith('text-davinci-002-render');
-    const { isUnofficialChatGptModel } = this;
-
-    // Davinci models have a max context length of 4097 tokens.
-    this.maxContextTokens = this.options.maxContextTokens || (isChatGptModel ? 4095 : 4097);
-    // I decided to reserve 1024 tokens for the response.
-    // The max prompt tokens is determined by the max context tokens minus the max response tokens.
-    // Earlier messages will be dropped until the prompt is within the limit.
-    this.maxResponseTokens = this.modelOptions.max_tokens || 1024;
-    this.maxPromptTokens =
-      this.options.maxPromptTokens || this.maxContextTokens - this.maxResponseTokens;
-
-    if (this.maxPromptTokens + this.maxResponseTokens > this.maxContextTokens) {
-      throw new Error(
-        `maxPromptTokens + max_tokens (${this.maxPromptTokens} + ${this.maxResponseTokens} = ${
-          this.maxPromptTokens + this.maxResponseTokens
-        }) must be less than or equal to maxContextTokens (${this.maxContextTokens})`,
-      );
-    }
-
-    this.userLabel = this.options.userLabel || 'User';
-    this.chatGptLabel = this.options.chatGptLabel || 'ChatGPT';
-
-    if (isChatGptModel) {
-      // Use these faux tokens to help the AI understand the context since we are building the chat log ourselves.
-      // Trying to use "<|im_start|>" causes the AI to still generate "<" or "<|" at the end sometimes for some reason,
-      // without tripping the stop sequences, so I'm using "||>" instead.
-      this.startToken = '||>';
-      this.endToken = '';
-      this.gptEncoder = this.constructor.getTokenizer('cl100k_base');
-    } else if (isUnofficialChatGptModel) {
-      this.startToken = '<|im_start|>';
-      this.endToken = '<|im_end|>';
-      this.gptEncoder = this.constructor.getTokenizer('text-davinci-003', true, {
-        '<|im_start|>': 100264,
-        '<|im_end|>': 100265,
-      });
-    } else {
-      // Previously I was trying to use "<|endoftext|>" but there seems to be some bug with OpenAI's token counting
-      // system that causes only the first "<|endoftext|>" to be counted as 1 token, and the rest are not treated
-      // as a single token. So we're using this instead.
-      this.startToken = '||>';
-      this.endToken = '';
-      try {
-        this.gptEncoder = this.constructor.getTokenizer(this.modelOptions.model, true);
-      } catch {
-        this.gptEncoder = this.constructor.getTokenizer('text-davinci-003', true);
-      }
-    }
-
-    if (!this.modelOptions.stop) {
-      const stopTokens = [this.startToken];
-      if (this.endToken && this.endToken !== this.startToken) {
-        stopTokens.push(this.endToken);
-      }
-      stopTokens.push(`\n${this.userLabel}:`);
-      stopTokens.push('<|diff_marker|>');
-      // I chose not to do one for `chatGptLabel` because I've never seen it happen
-      this.modelOptions.stop = stopTokens;
-    }
-
-    if (this.options.reverseProxyUrl) {
-      this.completionsUrl = this.options.reverseProxyUrl;
-    } else if (isChatGptModel) {
-      this.completionsUrl = 'https://api.openai.com/v1/chat/completions';
-    } else {
-      this.completionsUrl = 'https://api.openai.com/v1/completions';
-    }
-
-    return this;
-  }
-
-  static getTokenizer(encoding, isModelName = false, extendSpecialTokens = {}) {
-    if (tokenizersCache[encoding]) {
-      return tokenizersCache[encoding];
-    }
-    let tokenizer;
-    if (isModelName) {
-      tokenizer = encodingForModel(encoding, extendSpecialTokens);
-    } else {
-      tokenizer = getEncoding(encoding, extendSpecialTokens);
-    }
-    tokenizersCache[encoding] = tokenizer;
-    return tokenizer;
-  }
-
-  /** @type {getCompletion} */
-  async getCompletion(input, onProgress, onTokenProgress, abortController = null) {
-    if (!abortController) {
-      abortController = new AbortController();
-    }
-
-    let modelOptions = { ...this.modelOptions };
-    if (typeof onProgress === 'function') {
-      modelOptions.stream = true;
-    }
-    if (this.isChatGptModel) {
-      modelOptions.messages = input;
-    } else {
-      modelOptions.prompt = input;
-    }
-
-    if (this.useOpenRouter && modelOptions.prompt) {
-      delete modelOptions.stop;
-    }
-
-    const { debug } = this.options;
-    let baseURL = this.completionsUrl;
-    if (debug) {
-      console.debug();
-      console.debug(baseURL);
-      console.debug(modelOptions);
-      console.debug();
-    }
-
-    const opts = {
-      method: 'POST',
-      headers: {
-        'Content-Type': 'application/json',
-      },
-    };
-
-    if (this.isVisionModel) {
-      modelOptions.max_tokens = 4000;
-    }
-
-    /** @type {TAzureConfig | undefined} */
-    const azureConfig = this.options?.req?.app?.locals?.[EModelEndpoint.azureOpenAI];
-
-    const isAzure = this.azure || this.options.azure;
-    if (
-      (isAzure && this.isVisionModel && azureConfig) ||
-      (azureConfig && this.isVisionModel && this.options.endpoint === EModelEndpoint.azureOpenAI)
-    ) {
-      const { modelGroupMap, groupMap } = azureConfig;
-      const {
-        azureOptions,
-        baseURL,
-        headers = {},
-        serverless,
-      } = mapModelToAzureConfig({
-        modelName: modelOptions.model,
-        modelGroupMap,
-        groupMap,
-      });
-      opts.headers = resolveHeaders(headers);
-      this.langchainProxy = extractBaseURL(baseURL);
-      this.apiKey = azureOptions.azureOpenAIApiKey;
-
-      const groupName = modelGroupMap[modelOptions.model].group;
-      this.options.addParams = azureConfig.groupMap[groupName].addParams;
-      this.options.dropParams = azureConfig.groupMap[groupName].dropParams;
-      // Note: `forcePrompt` not re-assigned as only chat models are vision models
-
-      this.azure = !serverless && azureOptions;
-      this.azureEndpoint =
-        !serverless && genAzureChatCompletion(this.azure, modelOptions.model, this);
-      if (serverless === true) {
-        this.options.defaultQuery = azureOptions.azureOpenAIApiVersion
-          ? { 'api-version': azureOptions.azureOpenAIApiVersion }
-          : undefined;
-        this.options.headers['api-key'] = this.apiKey;
-      }
-    }
-
-    if (this.options.defaultQuery) {
-      opts.defaultQuery = this.options.defaultQuery;
-    }
-
-    if (this.options.headers) {
-      opts.headers = { ...opts.headers, ...this.options.headers };
-    }
-
-    if (isAzure) {
-      // Azure does not accept `model` in the body, so we need to remove it.
-      delete modelOptions.model;
-
-      baseURL = this.langchainProxy
-        ? constructAzureURL({
-            baseURL: this.langchainProxy,
-            azureOptions: this.azure,
-          })
-        : this.azureEndpoint.split(/(?<!\/)\/(chat|completion)\//)[0];
-
-      if (this.options.forcePrompt) {
-        baseURL += '/completions';
-      } else {
-        baseURL += '/chat/completions';
-      }
-
-      opts.defaultQuery = { 'api-version': this.azure.azureOpenAIApiVersion };
-      opts.headers = { ...opts.headers, 'api-key': this.apiKey };
-    } else if (this.apiKey) {
-      opts.headers.Authorization = `Bearer ${this.apiKey}`;
-    }
-
-    if (process.env.OPENAI_ORGANIZATION) {
-      opts.headers['OpenAI-Organization'] = process.env.OPENAI_ORGANIZATION;
-    }
-
-    if (this.useOpenRouter) {
-      opts.headers['HTTP-Referer'] = 'https://librechat.ai';
-      opts.headers['X-Title'] = 'LibreChat';
-    }
-
-    /* hacky fixes for Mistral AI API:
-      - Re-orders system message to the top of the messages payload, as not allowed anywhere else
-      - If there is only one message and it's a system message, change the role to user
-      */
-    if (baseURL.includes('https://api.mistral.ai/v1') && modelOptions.messages) {
-      const { messages } = modelOptions;
-
-      const systemMessageIndex = messages.findIndex((msg) => msg.role === 'system');
-
-      if (systemMessageIndex > 0) {
-        const [systemMessage] = messages.splice(systemMessageIndex, 1);
-        messages.unshift(systemMessage);
-      }
-
-      modelOptions.messages = messages;
-
-      if (messages.length === 1 && messages[0].role === 'system') {
-        modelOptions.messages[0].role = 'user';
-      }
-    }
-
-    if (this.options.addParams && typeof this.options.addParams === 'object') {
-      modelOptions = {
-        ...modelOptions,
-        ...this.options.addParams,
-      };
-      logger.debug('[ChatGPTClient] chatCompletion: added params', {
-        addParams: this.options.addParams,
-        modelOptions,
-      });
-    }
-
-    if (this.options.dropParams && Array.isArray(this.options.dropParams)) {
-      this.options.dropParams.forEach((param) => {
-        delete modelOptions[param];
-      });
-      logger.debug('[ChatGPTClient] chatCompletion: dropped params', {
-        dropParams: this.options.dropParams,
-        modelOptions,
-      });
-    }
-
-    if (baseURL.startsWith(CohereConstants.API_URL)) {
-      const payload = createCoherePayload({ modelOptions });
-      return await this.cohereChatCompletion({ payload, onTokenProgress });
-    }
-
-    if (baseURL.includes('v1') && !baseURL.includes('/completions') && !this.isChatCompletion) {
-      baseURL = baseURL.split('v1')[0] + 'v1/completions';
-    } else if (
-      baseURL.includes('v1') &&
-      !baseURL.includes('/chat/completions') &&
-      this.isChatCompletion
-    ) {
-      baseURL = baseURL.split('v1')[0] + 'v1/chat/completions';
-    }
-
-    const BASE_URL = new URL(baseURL);
-    if (opts.defaultQuery) {
-      Object.entries(opts.defaultQuery).forEach(([key, value]) => {
-        BASE_URL.searchParams.append(key, value);
-      });
-      delete opts.defaultQuery;
-    }
-
-    const completionsURL = BASE_URL.toString();
-    opts.body = JSON.stringify(modelOptions);
-
-    if (modelOptions.stream) {
-      return new Promise(async (resolve, reject) => {
-        try {
-          let done = false;
-          await fetchEventSource(completionsURL, {
-            ...opts,
-            signal: abortController.signal,
-            async onopen(response) {
-              if (response.status === 200) {
-                return;
-              }
-              if (debug) {
-                console.debug(response);
-              }
-              let error;
-              try {
-                const body = await response.text();
-                error = new Error(`Failed to send message. HTTP ${response.status} - ${body}`);
-                error.status = response.status;
-                error.json = JSON.parse(body);
-              } catch {
-                error = error || new Error(`Failed to send message. HTTP ${response.status}`);
-              }
-              throw error;
-            },
-            onclose() {
-              if (debug) {
-                console.debug('Server closed the connection unexpectedly, returning...');
-              }
-              // workaround for private API not sending [DONE] event
-              if (!done) {
-                onProgress('[DONE]');
-                resolve();
-              }
-            },
-            onerror(err) {
-              if (debug) {
-                console.debug(err);
-              }
-              // rethrow to stop the operation
-              throw err;
-            },
-            onmessage(message) {
-              if (debug) {
-                console.debug(message);
-              }
-              if (!message.data || message.event === 'ping') {
-                return;
-              }
-              if (message.data === '[DONE]') {
-                onProgress('[DONE]');
-                resolve();
-                done = true;
-                return;
-              }
-              onProgress(JSON.parse(message.data));
-            },
-          });
-        } catch (err) {
-          reject(err);
-        }
-      });
-    }
-    const response = await fetch(completionsURL, {
-      ...opts,
-      signal: abortController.signal,
-    });
-    if (response.status !== 200) {
-      const body = await response.text();
-      const error = new Error(`Failed to send message. HTTP ${response.status} - ${body}`);
-      error.status = response.status;
-      try {
-        error.json = JSON.parse(body);
-      } catch {
-        error.body = body;
-      }
-      throw error;
-    }
-    return response.json();
-  }
-
-  /** @type {cohereChatCompletion} */
-  async cohereChatCompletion({ payload, onTokenProgress }) {
-    const cohere = new CohereClient({
-      token: this.apiKey,
-      environment: this.completionsUrl,
-    });
-
-    if (!payload.stream) {
-      const chatResponse = await cohere.chat(payload);
-      return chatResponse.text;
-    }
-
-    const chatStream = await cohere.chatStream(payload);
-    let reply = '';
-    for await (const message of chatStream) {
-      if (!message) {
-        continue;
-      }
-
-      if (message.eventType === 'text-generation' && message.text) {
-        onTokenProgress(message.text);
-        reply += message.text;
-      }
-      /*
-      Cohere API Chinese Unicode character replacement hotfix.
-      Should be un-commented when the following issue is resolved:
-      https://github.com/cohere-ai/cohere-typescript/issues/151
-
-      else if (message.eventType === 'stream-end' && message.response) {
-        reply = message.response.text;
-      }
-      */
-    }
-
-    return reply;
-  }
-
-  async generateTitle(userMessage, botMessage) {
-    const instructionsPayload = {
-      role: 'system',
-      content: `Write an extremely concise subtitle for this conversation with no more than a few words. All words should be capitalized. Exclude punctuation.
-
-||>Message:
-${userMessage.message}
-||>Response:
-${botMessage.message}
-
-||>Title:`,
-    };
-
-    const titleGenClientOptions = JSON.parse(JSON.stringify(this.options));
-    titleGenClientOptions.modelOptions = {
-      model: 'gpt-3.5-turbo',
-      temperature: 0,
-      presence_penalty: 0,
-      frequency_penalty: 0,
-    };
-    const titleGenClient = new ChatGPTClient(this.apiKey, titleGenClientOptions);
-    const result = await titleGenClient.getCompletion([instructionsPayload], null);
-    // remove any non-alphanumeric characters, replace multiple spaces with 1, and then trim
-    return result.choices[0].message.content
-      .replace(/[^a-zA-Z0-9' ]/g, '')
-      .replace(/\s+/g, ' ')
-      .trim();
-  }
-
-  async sendMessage(message, opts = {}) {
-    if (opts.clientOptions && typeof opts.clientOptions === 'object') {
-      this.setOptions(opts.clientOptions);
-    }
-
-    const conversationId = opts.conversationId || crypto.randomUUID();
-    const parentMessageId = opts.parentMessageId || crypto.randomUUID();
-
-    let conversation =
-      typeof opts.conversation === 'object'
-        ? opts.conversation
-        : await this.conversationsCache.get(conversationId);
-
-    let isNewConversation = false;
-    if (!conversation) {
-      conversation = {
-        messages: [],
-        createdAt: Date.now(),
-      };
-      isNewConversation = true;
-    }
-
-    const shouldGenerateTitle = opts.shouldGenerateTitle && isNewConversation;
-
-    const userMessage = {
-      id: crypto.randomUUID(),
-      parentMessageId,
-      role: 'User',
-      message,
-    };
-    conversation.messages.push(userMessage);
-
-    // Doing it this way instead of having each message be a separate element in the array seems to be more reliable,
-    // especially when it comes to keeping the AI in character. It also seems to improve coherency and context retention.
-    const { prompt: payload, context } = await this.buildPrompt(
-      conversation.messages,
-      userMessage.id,
-      {
-        isChatGptModel: this.isChatGptModel,
-        promptPrefix: opts.promptPrefix,
-      },
-    );
-
-    if (this.options.keepNecessaryMessagesOnly) {
-      conversation.messages = context;
-    }
-
-    let reply = '';
-    let result = null;
-    if (typeof opts.onProgress === 'function') {
-      await this.getCompletion(
-        payload,
-        (progressMessage) => {
-          if (progressMessage === '[DONE]') {
-            return;
-          }
-          const token = this.isChatGptModel
-            ? progressMessage.choices[0].delta.content
-            : progressMessage.choices[0].text;
-          // first event's delta content is always undefined
-          if (!token) {
-            return;
-          }
-          if (this.options.debug) {
-            console.debug(token);
-          }
-          if (token === this.endToken) {
-            return;
-          }
-          opts.onProgress(token);
-          reply += token;
-        },
-        opts.abortController || new AbortController(),
-      );
-    } else {
-      result = await this.getCompletion(
-        payload,
-        null,
-        opts.abortController || new AbortController(),
-      );
-      if (this.options.debug) {
-        console.debug(JSON.stringify(result));
-      }
-      if (this.isChatGptModel) {
-        reply = result.choices[0].message.content;
-      } else {
-        reply = result.choices[0].text.replace(this.endToken, '');
-      }
-    }
-
-    // avoids some rendering issues when using the CLI app
-    if (this.options.debug) {
-      console.debug();
-    }
-
-    reply = reply.trim();
-
-    const replyMessage = {
-      id: crypto.randomUUID(),
-      parentMessageId: userMessage.id,
-      role: 'ChatGPT',
-      message: reply,
-    };
-    conversation.messages.push(replyMessage);
-
-    const returnData = {
-      response: replyMessage.message,
-      conversationId,
-      parentMessageId: replyMessage.parentMessageId,
-      messageId: replyMessage.id,
-      details: result || {},
-    };
-
-    if (shouldGenerateTitle) {
-      conversation.title = await this.generateTitle(userMessage, replyMessage);
-      returnData.title = conversation.title;
-    }
-
-    await this.conversationsCache.set(conversationId, conversation);
-
-    if (this.options.returnConversation) {
-      returnData.conversation = conversation;
-    }
-
-    return returnData;
-  }
-
-  async buildPrompt(messages, { isChatGptModel = false, promptPrefix = null }) {
-    promptPrefix = (promptPrefix || this.options.promptPrefix || '').trim();
-
-    // Handle attachments and create augmentedPrompt
-    if (this.options.attachments) {
-      const attachments = await this.options.attachments;
-      const lastMessage = messages[messages.length - 1];
-
-      if (this.message_file_map) {
-        this.message_file_map[lastMessage.messageId] = attachments;
-      } else {
-        this.message_file_map = {
-          [lastMessage.messageId]: attachments,
-        };
-      }
-
-      const files = await this.addImageURLs(lastMessage, attachments);
-      this.options.attachments = files;
-
-      this.contextHandlers = createContextHandlers(this.options.req, lastMessage.text);
-    }
-
-    if (this.message_file_map) {
-      this.contextHandlers = createContextHandlers(
-        this.options.req,
-        messages[messages.length - 1].text,
-      );
-    }
-
-    // Calculate image token cost and process embedded files
-    messages.forEach((message, i) => {
-      if (this.message_file_map && this.message_file_map[message.messageId]) {
-        const attachments = this.message_file_map[message.messageId];
-        for (const file of attachments) {
-          if (file.embedded) {
-            this.contextHandlers?.processFile(file);
-            continue;
-          }
-
-          messages[i].tokenCount =
-            (messages[i].tokenCount || 0) +
-            this.calculateImageTokenCost({
-              width: file.width,
-              height: file.height,
-              detail: this.options.imageDetail ?? ImageDetail.auto,
-            });
-        }
-      }
-    });
-
-    if (this.contextHandlers) {
-      this.augmentedPrompt = await this.contextHandlers.createContext();
-      promptPrefix = this.augmentedPrompt + promptPrefix;
-    }
-
-    if (promptPrefix) {
-      // If the prompt prefix doesn't end with the end token, add it.
-      if (!promptPrefix.endsWith(`${this.endToken}`)) {
-        promptPrefix = `${promptPrefix.trim()}${this.endToken}\n\n`;
-      }
-      promptPrefix = `${this.startToken}Instructions:\n${promptPrefix}`;
-    }
-    const promptSuffix = `${this.startToken}${this.chatGptLabel}:\n`; // Prompt ChatGPT to respond.
-
-    const instructionsPayload = {
-      role: 'system',
-      content: promptPrefix,
-    };
-
-    const messagePayload = {
-      role: 'system',
-      content: promptSuffix,
-    };
-
-    let currentTokenCount;
-    if (isChatGptModel) {
-      currentTokenCount =
-        this.getTokenCountForMessage(instructionsPayload) +
-        this.getTokenCountForMessage(messagePayload);
-    } else {
-      currentTokenCount = this.getTokenCount(`${promptPrefix}${promptSuffix}`);
-    }
-    let promptBody = '';
-    const maxTokenCount = this.maxPromptTokens;
-
-    const context = [];
-
-    // Iterate backwards through the messages, adding them to the prompt until we reach the max token count.
-    // Do this within a recursive async function so that it doesn't block the event loop for too long.
-    const buildPromptBody = async () => {
-      if (currentTokenCount < maxTokenCount && messages.length > 0) {
-        const message = messages.pop();
-        const roleLabel =
-          message?.isCreatedByUser || message?.role?.toLowerCase() === 'user'
-            ? this.userLabel
-            : this.chatGptLabel;
-        const messageString = `${this.startToken}${roleLabel}:\n${
-          message?.text ?? message?.message
-        }${this.endToken}\n`;
-        let newPromptBody;
-        if (promptBody || isChatGptModel) {
-          newPromptBody = `${messageString}${promptBody}`;
-        } else {
-          // Always insert prompt prefix before the last user message, if not gpt-3.5-turbo.
-          // This makes the AI obey the prompt instructions better, which is important for custom instructions.
-          // After a bunch of testing, it doesn't seem to cause the AI any confusion, even if you ask it things
-          // like "what's the last thing I wrote?".
-          newPromptBody = `${promptPrefix}${messageString}${promptBody}`;
-        }
-
-        context.unshift(message);
-
-        const tokenCountForMessage = this.getTokenCount(messageString);
-        const newTokenCount = currentTokenCount + tokenCountForMessage;
-        if (newTokenCount > maxTokenCount) {
-          if (promptBody) {
-            // This message would put us over the token limit, so don't add it.
-            return false;
-          }
-          // This is the first message, so we can't add it. Just throw an error.
-          throw new Error(
-            `Prompt is too long. Max token count is ${maxTokenCount}, but prompt is ${newTokenCount} tokens long.`,
-          );
-        }
-        promptBody = newPromptBody;
-        currentTokenCount = newTokenCount;
-        // wait for next tick to avoid blocking the event loop
-        await new Promise((resolve) => setImmediate(resolve));
-        return buildPromptBody();
-      }
-      return true;
-    };
-
-    await buildPromptBody();
-
-    const prompt = `${promptBody}${promptSuffix}`;
-    if (isChatGptModel) {
-      messagePayload.content = prompt;
-      // Add 3 tokens for Assistant Label priming after all messages have been counted.
-      currentTokenCount += 3;
-    }
-
-    // Use up to `this.maxContextTokens` tokens (prompt + response), but try to leave `this.maxTokens` tokens for the response.
-    this.modelOptions.max_tokens = Math.min(
-      this.maxContextTokens - currentTokenCount,
-      this.maxResponseTokens,
-    );
-
-    if (isChatGptModel) {
-      return { prompt: [instructionsPayload, messagePayload], context };
-    }
-    return { prompt, context, promptTokens: currentTokenCount };
-  }
-
-  getTokenCount(text) {
-    return this.gptEncoder.encode(text, 'all').length;
-  }
-
-  /**
-   * Algorithm adapted from "6. Counting tokens for chat API calls" of
-   * https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb
-   *
-   * An additional 3 tokens need to be added for assistant label priming after all messages have been counted.
-   *
-   * @param {Object} message
-   */
-  getTokenCountForMessage(message) {
-    // Note: gpt-3.5-turbo and gpt-4 may update over time. Use default for these as well as for unknown models
-    let tokensPerMessage = 3;
-    let tokensPerName = 1;
-
-    if (this.modelOptions.model === 'gpt-3.5-turbo-0301') {
-      tokensPerMessage = 4;
-      tokensPerName = -1;
-    }
-
-    let numTokens = tokensPerMessage;
-    for (let [key, value] of Object.entries(message)) {
-      numTokens += this.getTokenCount(value);
-      if (key === 'name') {
-        numTokens += tokensPerName;
-      }
-    }
-
-    return numTokens;
-  }
-}
-
-module.exports = ChatGPTClient;

api/app/clients/GoogleClient.js

@@ -1,7 +1,8 @@
 const { google } = require('googleapis');
-const { Tokenizer } = require('@librechat/api');
+const { getModelMaxTokens } = require('@librechat/api');
 const { concat } = require('@langchain/core/utils/stream');
 const { ChatVertexAI } = require('@langchain/google-vertexai');
+const { Tokenizer, getSafetySettings } = require('@librechat/api');
 const { ChatGoogleGenerativeAI } = require('@langchain/google-genai');
 const { GoogleGenerativeAI: GenAI } = require('@google/generative-ai');
 const { HumanMessage, SystemMessage } = require('@langchain/core/messages');
@@ -12,16 +13,15 @@ const {
   endpointSettings,
   parseTextParts,
   EModelEndpoint,
+  googleSettings,
   ContentTypes,
   VisionModes,
   ErrorTypes,
   Constants,
   AuthKeys,
 } = require('librechat-data-provider');
-const { getSafetySettings } = require('~/server/services/Endpoints/google/llm');
 const { encodeAndFormat } = require('~/server/services/Files/images');
 const { spendTokens } = require('~/models/spendTokens');
-const { getModelMaxTokens } = require('~/utils');
 const { sleep } = require('~/server/utils');
 const { logger } = require('~/config');
 const {
@@ -166,6 +166,16 @@ class GoogleClient extends BaseClient {
       );
     }
 
+    // Add thinking configuration
+    this.modelOptions.thinkingConfig = {
+      thinkingBudget:
+        (this.modelOptions.thinking ?? googleSettings.thinking.default)
+          ? this.modelOptions.thinkingBudget
+          : 0,
+    };
+    delete this.modelOptions.thinking;
+    delete this.modelOptions.thinkingBudget;
+
     this.sender =
       this.options.sender ??
       getResponseSender({

api/app/clients/OpenAIClient.js

@@ -5,8 +5,11 @@ const {
   isEnabled,
   Tokenizer,
   createFetch,
+  resolveHeaders,
   constructAzureURL,
+  getModelMaxTokens,
   genAzureChatCompletion,
+  getModelMaxOutputTokens,
   createStreamEventHandlers,
 } = require('@librechat/api');
 const {
@@ -15,7 +18,6 @@ const {
   ContentTypes,
   parseTextParts,
   EModelEndpoint,
-  resolveHeaders,
   KnownEndpoints,
   openAISettings,
   ImageDetailCost,
@@ -31,28 +33,21 @@ const {
   titleInstruction,
   createContextHandlers,
 } = require('./prompts');
-const { extractBaseURL, getModelMaxTokens, getModelMaxOutputTokens } = require('~/utils');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
 const { addSpaceIfNeeded, sleep } = require('~/server/utils');
 const { spendTokens } = require('~/models/spendTokens');
 const { handleOpenAIErrors } = require('./tools/util');
-const { createLLM, RunManager } = require('./llm');
-const ChatGPTClient = require('./ChatGPTClient');
 const { summaryBuffer } = require('./memory');
 const { runTitleChain } = require('./chains');
+const { extractBaseURL } = require('~/utils');
 const { tokenSplit } = require('./document');
 const BaseClient = require('./BaseClient');
+const { createLLM } = require('./llm');
 const { logger } = require('~/config');
 
 class OpenAIClient extends BaseClient {
   constructor(apiKey, options = {}) {
     super(apiKey, options);
-    this.ChatGPTClient = new ChatGPTClient();
-    this.buildPrompt = this.ChatGPTClient.buildPrompt.bind(this);
-    /** @type {getCompletion} */
-    this.getCompletion = this.ChatGPTClient.getCompletion.bind(this);
-    /** @type {cohereChatCompletion} */
-    this.cohereChatCompletion = this.ChatGPTClient.cohereChatCompletion.bind(this);
     this.contextStrategy = options.contextStrategy
       ? options.contextStrategy.toLowerCase()
       : 'discard';
@@ -379,23 +374,12 @@ class OpenAIClient extends BaseClient {
     return files;
   }
 
-  async buildMessages(
-    messages,
-    parentMessageId,
-    { isChatCompletion = false, promptPrefix = null },
-    opts,
-  ) {
+  async buildMessages(messages, parentMessageId, { promptPrefix = null }, opts) {
     let orderedMessages = this.constructor.getMessagesForConversation({
       messages,
       parentMessageId,
       summary: this.shouldSummarize,
     });
-    if (!isChatCompletion) {
-      return await this.buildPrompt(orderedMessages, {
-        isChatGptModel: isChatCompletion,
-        promptPrefix,
-      });
-    }
 
     let payload;
     let instructions;
@@ -636,10 +620,6 @@ class OpenAIClient extends BaseClient {
     temperature = 0.2,
     max_tokens,
     streaming,
-    context,
-    tokenBuffer,
-    initialMessageCount,
-    conversationId,
   }) {
     const modelOptions = {
       modelName: modelName ?? model,
@@ -671,8 +651,10 @@ class OpenAIClient extends BaseClient {
     if (headers && typeof headers === 'object' && !Array.isArray(headers)) {
       configOptions.baseOptions = {
         headers: resolveHeaders({
-          ...headers,
-          ...configOptions?.baseOptions?.headers,
+          headers: {
+            ...headers,
+            ...configOptions?.baseOptions?.headers,
+          },
         }),
       };
     }
@@ -682,22 +664,12 @@ class OpenAIClient extends BaseClient {
       configOptions.httpsAgent = new HttpsProxyAgent(this.options.proxy);
     }
 
-    const { req, res, debug } = this.options;
-    const runManager = new RunManager({ req, res, debug, abortController: this.abortController });
-    this.runManager = runManager;
-
     const llm = createLLM({
       modelOptions,
       configOptions,
       openAIApiKey: this.apiKey,
       azure: this.azure,
       streaming,
-      callbacks: runManager.createCallbacks({
-        context,
-        tokenBuffer,
-        conversationId: this.conversationId ?? conversationId,
-        initialMessageCount,
-      }),
     });
 
     return llm;
@@ -718,6 +690,7 @@ class OpenAIClient extends BaseClient {
    *                            In case of failure, it will return the default title, "New Chat".
    */
   async titleConvo({ text, conversationId, responseText = '' }) {
+    const appConfig = this.options.req?.config;
     this.conversationId = conversationId;
 
     if (this.options.attachments) {
@@ -746,8 +719,7 @@ class OpenAIClient extends BaseClient {
       max_tokens: 16,
     };
 
-    /** @type {TAzureConfig | undefined} */
-    const azureConfig = this.options?.req?.app?.locals?.[EModelEndpoint.azureOpenAI];
+    const azureConfig = appConfig?.endpoints?.[EModelEndpoint.azureOpenAI];
 
     const resetTitleOptions = !!(
       (this.azure && azureConfig) ||
@@ -767,7 +739,7 @@ class OpenAIClient extends BaseClient {
         groupMap,
       });
 
-      this.options.headers = resolveHeaders(headers);
+      this.options.headers = resolveHeaders({ headers });
       this.options.reverseProxyUrl = baseURL ?? null;
       this.langchainProxy = extractBaseURL(this.options.reverseProxyUrl);
       this.apiKey = azureOptions.azureOpenAIApiKey;
@@ -1136,6 +1108,7 @@ ${convo}
   }
 
   async chatCompletion({ payload, onProgress, abortController = null }) {
+    const appConfig = this.options.req?.config;
     let error = null;
     let intermediateReply = [];
     const errorCallback = (err) => (error = err);
@@ -1159,6 +1132,7 @@ ${convo}
       logger.debug('[OpenAIClient] chatCompletion', { baseURL, modelOptions });
       const opts = {
         baseURL,
+        fetchOptions: {},
       };
 
       if (this.useOpenRouter) {
@@ -1177,11 +1151,10 @@ ${convo}
       }
 
       if (this.options.proxy) {
-        opts.httpAgent = new HttpsProxyAgent(this.options.proxy);
+        opts.fetchOptions.agent = new HttpsProxyAgent(this.options.proxy);
       }
 
-      /** @type {TAzureConfig | undefined} */
-      const azureConfig = this.options?.req?.app?.locals?.[EModelEndpoint.azureOpenAI];
+      const azureConfig = appConfig?.endpoints?.[EModelEndpoint.azureOpenAI];
 
       if (
         (this.azure && this.isVisionModel && azureConfig) ||
@@ -1198,7 +1171,7 @@ ${convo}
           modelGroupMap,
           groupMap,
         });
-        opts.defaultHeaders = resolveHeaders(headers);
+        opts.defaultHeaders = resolveHeaders({ headers });
         this.langchainProxy = extractBaseURL(baseURL);
         this.apiKey = azureOptions.azureOpenAIApiKey;
 
@@ -1239,7 +1212,9 @@ ${convo}
       }
 
       if (this.isOmni === true && modelOptions.max_tokens != null) {
-        modelOptions.max_completion_tokens = modelOptions.max_tokens;
+        const paramName =
+          modelOptions.useResponsesApi === true ? 'max_output_tokens' : 'max_completion_tokens';
+        modelOptions[paramName] = modelOptions.max_tokens;
         delete modelOptions.max_tokens;
       }
       if (this.isOmni === true && modelOptions.temperature != null) {
@@ -1395,7 +1370,7 @@ ${convo}
           ...modelOptions,
           stream: true,
         };
-        const stream = await openai.beta.chat.completions
+        const stream = await openai.chat.completions
           .stream(params)
           .on('abort', () => {
             /* Do nothing here */

api/app/clients/PluginsClient.js

@@ -1,542 +0,0 @@
-const OpenAIClient = require('./OpenAIClient');
-const { CallbackManager } = require('@langchain/core/callbacks/manager');
-const { BufferMemory, ChatMessageHistory } = require('langchain/memory');
-const { addImages, buildErrorInput, buildPromptPrefix } = require('./output_parsers');
-const { initializeCustomAgent, initializeFunctionsAgent } = require('./agents');
-const { processFileURL } = require('~/server/services/Files/process');
-const { EModelEndpoint } = require('librechat-data-provider');
-const { checkBalance } = require('~/models/balanceMethods');
-const { formatLangChainMessages } = require('./prompts');
-const { extractBaseURL } = require('~/utils');
-const { loadTools } = require('./tools/util');
-const { logger } = require('~/config');
-
-class PluginsClient extends OpenAIClient {
-  constructor(apiKey, options = {}) {
-    super(apiKey, options);
-    this.sender = options.sender ?? 'Assistant';
-    this.tools = [];
-    this.actions = [];
-    this.setOptions(options);
-    this.openAIApiKey = this.apiKey;
-    this.executor = null;
-  }
-
-  setOptions(options) {
-    this.agentOptions = { ...options.agentOptions };
-    this.functionsAgent = this.agentOptions?.agent === 'functions';
-    this.agentIsGpt3 = this.agentOptions?.model?.includes('gpt-3');
-
-    super.setOptions(options);
-
-    this.isGpt3 = this.modelOptions?.model?.includes('gpt-3');
-
-    if (this.options.reverseProxyUrl) {
-      this.langchainProxy = extractBaseURL(this.options.reverseProxyUrl);
-    }
-  }
-
-  getSaveOptions() {
-    return {
-      artifacts: this.options.artifacts,
-      chatGptLabel: this.options.chatGptLabel,
-      modelLabel: this.options.modelLabel,
-      promptPrefix: this.options.promptPrefix,
-      tools: this.options.tools,
-      ...this.modelOptions,
-      agentOptions: this.agentOptions,
-      iconURL: this.options.iconURL,
-      greeting: this.options.greeting,
-      spec: this.options.spec,
-    };
-  }
-
-  saveLatestAction(action) {
-    this.actions.push(action);
-  }
-
-  getFunctionModelName(input) {
-    if (/-(?!0314)\d{4}/.test(input)) {
-      return input;
-    } else if (input.includes('gpt-3.5-turbo')) {
-      return 'gpt-3.5-turbo';
-    } else if (input.includes('gpt-4')) {
-      return 'gpt-4';
-    } else {
-      return 'gpt-3.5-turbo';
-    }
-  }
-
-  getBuildMessagesOptions(opts) {
-    return {
-      isChatCompletion: true,
-      promptPrefix: opts.promptPrefix,
-      abortController: opts.abortController,
-    };
-  }
-
-  async initialize({ user, message, onAgentAction, onChainEnd, signal }) {
-    const modelOptions = {
-      modelName: this.agentOptions.model,
-      temperature: this.agentOptions.temperature,
-    };
-
-    const model = this.initializeLLM({
-      ...modelOptions,
-      context: 'plugins',
-      initialMessageCount: this.currentMessages.length + 1,
-    });
-
-    logger.debug(
-      `[PluginsClient] Agent Model: ${model.modelName} | Temp: ${model.temperature} | Functions: ${this.functionsAgent}`,
-    );
-
-    // Map Messages to Langchain format
-    const pastMessages = formatLangChainMessages(this.currentMessages.slice(0, -1), {
-      userName: this.options?.name,
-    });
-    logger.debug('[PluginsClient] pastMessages: ' + pastMessages.length);
-
-    // TODO: use readOnly memory, TokenBufferMemory? (both unavailable in LangChainJS)
-    const memory = new BufferMemory({
-      llm: model,
-      chatHistory: new ChatMessageHistory(pastMessages),
-    });
-
-    const { loadedTools } = await loadTools({
-      user,
-      model,
-      tools: this.options.tools,
-      functions: this.functionsAgent,
-      options: {
-        memory,
-        signal: this.abortController.signal,
-        openAIApiKey: this.openAIApiKey,
-        conversationId: this.conversationId,
-        fileStrategy: this.options.req.app.locals.fileStrategy,
-        processFileURL,
-        message,
-      },
-      useSpecs: true,
-    });
-
-    if (loadedTools.length === 0) {
-      return;
-    }
-
-    this.tools = loadedTools;
-
-    logger.debug('[PluginsClient] Requested Tools', this.options.tools);
-    logger.debug(
-      '[PluginsClient] Loaded Tools',
-      this.tools.map((tool) => tool.name),
-    );
-
-    const handleAction = (action, runId, callback = null) => {
-      this.saveLatestAction(action);
-
-      logger.debug('[PluginsClient] Latest Agent Action ', this.actions[this.actions.length - 1]);
-
-      if (typeof callback === 'function') {
-        callback(action, runId);
-      }
-    };
-
-    // initialize agent
-    const initializer = this.functionsAgent ? initializeFunctionsAgent : initializeCustomAgent;
-
-    let customInstructions = (this.options.promptPrefix ?? '').trim();
-    if (typeof this.options.artifactsPrompt === 'string' && this.options.artifactsPrompt) {
-      customInstructions = `${customInstructions ?? ''}\n${this.options.artifactsPrompt}`.trim();
-    }
-
-    this.executor = await initializer({
-      model,
-      signal,
-      pastMessages,
-      tools: this.tools,
-      customInstructions,
-      verbose: this.options.debug,
-      returnIntermediateSteps: true,
-      customName: this.options.chatGptLabel,
-      currentDateString: this.currentDateString,
-      callbackManager: CallbackManager.fromHandlers({
-        async handleAgentAction(action, runId) {
-          handleAction(action, runId, onAgentAction);
-        },
-        async handleChainEnd(action) {
-          if (typeof onChainEnd === 'function') {
-            onChainEnd(action);
-          }
-        },
-      }),
-    });
-
-    logger.debug('[PluginsClient] Loaded agent.');
-  }
-
-  async executorCall(message, { signal, stream, onToolStart, onToolEnd }) {
-    let errorMessage = '';
-    const maxAttempts = 1;
-
-    for (let attempts = 1; attempts <= maxAttempts; attempts++) {
-      const errorInput = buildErrorInput({
-        message,
-        errorMessage,
-        actions: this.actions,
-        functionsAgent: this.functionsAgent,
-      });
-      const input = attempts > 1 ? errorInput : message;
-
-      logger.debug(`[PluginsClient] Attempt ${attempts} of ${maxAttempts}`);
-
-      if (errorMessage.length > 0) {
-        logger.debug('[PluginsClient] Caught error, input: ' + JSON.stringify(input));
-      }
-
-      try {
-        this.result = await this.executor.call({ input, signal }, [
-          {
-            async handleToolStart(...args) {
-              await onToolStart(...args);
-            },
-            async handleToolEnd(...args) {
-              await onToolEnd(...args);
-            },
-            async handleLLMEnd(output) {
-              const { generations } = output;
-              const { text } = generations[0][0];
-              if (text && typeof stream === 'function') {
-                await stream(text);
-              }
-            },
-          },
-        ]);
-        break; // Exit the loop if the function call is successful
-      } catch (err) {
-        logger.error('[PluginsClient] executorCall error:', err);
-        if (attempts === maxAttempts) {
-          const { run } = this.runManager.getRunByConversationId(this.conversationId);
-          const defaultOutput = `Encountered an error while attempting to respond: ${err.message}`;
-          this.result.output = run && run.error ? run.error : defaultOutput;
-          this.result.errorMessage = run && run.error ? run.error : err.message;
-          this.result.intermediateSteps = this.actions;
-          break;
-        }
-      }
-    }
-  }
-
-  /**
-   *
-   * @param {TMessage} responseMessage
-   * @param {Partial<TMessage>} saveOptions
-   * @param {string} user
-   * @returns
-   */
-  async handleResponseMessage(responseMessage, saveOptions, user) {
-    const { output, errorMessage, ...result } = this.result;
-    logger.debug('[PluginsClient][handleResponseMessage] Output:', {
-      output,
-      errorMessage,
-      ...result,
-    });
-    const { error } = responseMessage;
-    if (!error) {
-      responseMessage.tokenCount = this.getTokenCountForResponse(responseMessage);
-      responseMessage.completionTokens = this.getTokenCount(responseMessage.text);
-    }
-
-    // Record usage only when completion is skipped as it is already recorded in the agent phase.
-    if (!this.agentOptions.skipCompletion && !error) {
-      await this.recordTokenUsage(responseMessage);
-    }
-
-    const databasePromise = this.saveMessageToDatabase(responseMessage, saveOptions, user);
-    delete responseMessage.tokenCount;
-    return { ...responseMessage, ...result, databasePromise };
-  }
-
-  async sendMessage(message, opts = {}) {
-    /** @type {Promise<TMessage>} */
-    let userMessagePromise;
-    /** @type {{ filteredTools: string[], includedTools: string[] }} */
-    const { filteredTools = [], includedTools = [] } = this.options.req.app.locals;
-
-    if (includedTools.length > 0) {
-      const tools = this.options.tools.filter((plugin) => includedTools.includes(plugin));
-      this.options.tools = tools;
-    } else {
-      const tools = this.options.tools.filter((plugin) => !filteredTools.includes(plugin));
-      this.options.tools = tools;
-    }
-
-    // If a message is edited, no tools can be used.
-    const completionMode = this.options.tools.length === 0 || opts.isEdited;
-    if (completionMode) {
-      this.setOptions(opts);
-      return super.sendMessage(message, opts);
-    }
-
-    logger.debug('[PluginsClient] sendMessage', { userMessageText: message, opts });
-    const {
-      user,
-      conversationId,
-      responseMessageId,
-      saveOptions,
-      userMessage,
-      onAgentAction,
-      onChainEnd,
-      onToolStart,
-      onToolEnd,
-    } = await this.handleStartMethods(message, opts);
-
-    if (opts.progressCallback) {
-      opts.onProgress = opts.progressCallback.call(null, {
-        ...(opts.progressOptions ?? {}),
-        parentMessageId: userMessage.messageId,
-        messageId: responseMessageId,
-      });
-    }
-
-    this.currentMessages.push(userMessage);
-
-    let {
-      prompt: payload,
-      tokenCountMap,
-      promptTokens,
-    } = await this.buildMessages(
-      this.currentMessages,
-      userMessage.messageId,
-      this.getBuildMessagesOptions({
-        promptPrefix: null,
-        abortController: this.abortController,
-      }),
-    );
-
-    if (tokenCountMap) {
-      logger.debug('[PluginsClient] tokenCountMap', { tokenCountMap });
-      if (tokenCountMap[userMessage.messageId]) {
-        userMessage.tokenCount = tokenCountMap[userMessage.messageId];
-        logger.debug('[PluginsClient] userMessage.tokenCount', userMessage.tokenCount);
-      }
-      this.handleTokenCountMap(tokenCountMap);
-    }
-
-    this.result = {};
-    if (payload) {
-      this.currentMessages = payload;
-    }
-
-    if (!this.skipSaveUserMessage) {
-      userMessagePromise = this.saveMessageToDatabase(userMessage, saveOptions, user);
-      if (typeof opts?.getReqData === 'function') {
-        opts.getReqData({
-          userMessagePromise,
-        });
-      }
-    }
-
-    const balance = this.options.req?.app?.locals?.balance;
-    if (balance?.enabled) {
-      await checkBalance({
-        req: this.options.req,
-        res: this.options.res,
-        txData: {
-          user: this.user,
-          tokenType: 'prompt',
-          amount: promptTokens,
-          debug: this.options.debug,
-          model: this.modelOptions.model,
-          endpoint: EModelEndpoint.openAI,
-        },
-      });
-    }
-
-    const responseMessage = {
-      endpoint: EModelEndpoint.gptPlugins,
-      iconURL: this.options.iconURL,
-      messageId: responseMessageId,
-      conversationId,
-      parentMessageId: userMessage.messageId,
-      isCreatedByUser: false,
-      model: this.modelOptions.model,
-      sender: this.sender,
-      promptTokens,
-    };
-
-    await this.initialize({
-      user,
-      message,
-      onAgentAction,
-      onChainEnd,
-      signal: this.abortController.signal,
-      onProgress: opts.onProgress,
-    });
-
-    // const stream = async (text) => {
-    //   await this.generateTextStream.call(this, text, opts.onProgress, { delay: 1 });
-    // };
-    await this.executorCall(message, {
-      signal: this.abortController.signal,
-      // stream,
-      onToolStart,
-      onToolEnd,
-    });
-
-    // If message was aborted mid-generation
-    if (this.result?.errorMessage?.length > 0 && this.result?.errorMessage?.includes('cancel')) {
-      responseMessage.text = 'Cancelled.';
-      return await this.handleResponseMessage(responseMessage, saveOptions, user);
-    }
-
-    // If error occurred during generation (likely token_balance)
-    if (this.result?.errorMessage?.length > 0) {
-      responseMessage.error = true;
-      responseMessage.text = this.result.output;
-      return await this.handleResponseMessage(responseMessage, saveOptions, user);
-    }
-
-    if (this.agentOptions.skipCompletion && this.result.output && this.functionsAgent) {
-      const partialText = opts.getPartialText();
-      const trimmedPartial = opts.getPartialText().replaceAll(':::plugin:::\n', '');
-      responseMessage.text =
-        trimmedPartial.length === 0 ? `${partialText}${this.result.output}` : partialText;
-      addImages(this.result.intermediateSteps, responseMessage);
-      await this.generateTextStream(this.result.output, opts.onProgress, { delay: 5 });
-      return await this.handleResponseMessage(responseMessage, saveOptions, user);
-    }
-
-    if (this.agentOptions.skipCompletion && this.result.output) {
-      responseMessage.text = this.result.output;
-      addImages(this.result.intermediateSteps, responseMessage);
-      await this.generateTextStream(this.result.output, opts.onProgress, { delay: 5 });
-      return await this.handleResponseMessage(responseMessage, saveOptions, user);
-    }
-
-    logger.debug('[PluginsClient] Completion phase: this.result', this.result);
-
-    const promptPrefix = buildPromptPrefix({
-      result: this.result,
-      message,
-      functionsAgent: this.functionsAgent,
-    });
-
-    logger.debug('[PluginsClient]', { promptPrefix });
-
-    payload = await this.buildCompletionPrompt({
-      messages: this.currentMessages,
-      promptPrefix,
-    });
-
-    logger.debug('[PluginsClient] buildCompletionPrompt Payload', payload);
-    responseMessage.text = await this.sendCompletion(payload, opts);
-    return await this.handleResponseMessage(responseMessage, saveOptions, user);
-  }
-
-  async buildCompletionPrompt({ messages, promptPrefix: _promptPrefix }) {
-    logger.debug('[PluginsClient] buildCompletionPrompt messages', messages);
-
-    const orderedMessages = messages;
-    let promptPrefix = _promptPrefix.trim();
-    // If the prompt prefix doesn't end with the end token, add it.
-    if (!promptPrefix.endsWith(`${this.endToken}`)) {
-      promptPrefix = `${promptPrefix.trim()}${this.endToken}\n\n`;
-    }
-    promptPrefix = `${this.startToken}Instructions:\n${promptPrefix}`;
-    const promptSuffix = `${this.startToken}${this.chatGptLabel ?? 'Assistant'}:\n`;
-
-    const instructionsPayload = {
-      role: 'system',
-      content: promptPrefix,
-    };
-
-    const messagePayload = {
-      role: 'system',
-      content: promptSuffix,
-    };
-
-    if (this.isGpt3) {
-      instructionsPayload.role = 'user';
-      messagePayload.role = 'user';
-      instructionsPayload.content += `\n${promptSuffix}`;
-    }
-
-    // testing if this works with browser endpoint
-    if (!this.isGpt3 && this.options.reverseProxyUrl) {
-      instructionsPayload.role = 'user';
-    }
-
-    let currentTokenCount =
-      this.getTokenCountForMessage(instructionsPayload) +
-      this.getTokenCountForMessage(messagePayload);
-
-    let promptBody = '';
-    const maxTokenCount = this.maxPromptTokens;
-    // Iterate backwards through the messages, adding them to the prompt until we reach the max token count.
-    // Do this within a recursive async function so that it doesn't block the event loop for too long.
-    const buildPromptBody = async () => {
-      if (currentTokenCount < maxTokenCount && orderedMessages.length > 0) {
-        const message = orderedMessages.pop();
-        const isCreatedByUser = message.isCreatedByUser || message.role?.toLowerCase() === 'user';
-        const roleLabel = isCreatedByUser ? this.userLabel : this.chatGptLabel;
-        let messageString = `${this.startToken}${roleLabel}:\n${
-          message.text ?? message.content ?? ''
-        }${this.endToken}\n`;
-        let newPromptBody = `${messageString}${promptBody}`;
-
-        const tokenCountForMessage = this.getTokenCount(messageString);
-        const newTokenCount = currentTokenCount + tokenCountForMessage;
-        if (newTokenCount > maxTokenCount) {
-          if (promptBody) {
-            // This message would put us over the token limit, so don't add it.
-            return false;
-          }
-          // This is the first message, so we can't add it. Just throw an error.
-          throw new Error(
-            `Prompt is too long. Max token count is ${maxTokenCount}, but prompt is ${newTokenCount} tokens long.`,
-          );
-        }
-        promptBody = newPromptBody;
-        currentTokenCount = newTokenCount;
-        // wait for next tick to avoid blocking the event loop
-        await new Promise((resolve) => setTimeout(resolve, 0));
-        return buildPromptBody();
-      }
-      return true;
-    };
-
-    await buildPromptBody();
-    const prompt = promptBody;
-    messagePayload.content = prompt;
-    // Add 2 tokens for metadata after all messages have been counted.
-    currentTokenCount += 2;
-
-    if (this.isGpt3 && messagePayload.content.length > 0) {
-      const context = 'Chat History:\n';
-      messagePayload.content = `${context}${prompt}`;
-      currentTokenCount += this.getTokenCount(context);
-    }
-
-    // Use up to `this.maxContextTokens` tokens (prompt + response), but try to leave `this.maxTokens` tokens for the response.
-    this.modelOptions.max_tokens = Math.min(
-      this.maxContextTokens - currentTokenCount,
-      this.maxResponseTokens,
-    );
-
-    if (this.isGpt3) {
-      messagePayload.content += promptSuffix;
-      return [instructionsPayload, messagePayload];
-    }
-
-    const result = [messagePayload, instructionsPayload];
-
-    if (this.functionsAgent && !this.isGpt3) {
-      result[1].content = `${result[1].content}\n${this.startToken}${this.chatGptLabel}:\nSure thing! Here is the output you requested:\n`;
-    }
-
-    return result.filter((message) => message.content.length > 0);
-  }
-}
-
-module.exports = PluginsClient;

api/app/clients/callbacks/createStartHandler.js

@@ -1,95 +0,0 @@
-const { promptTokensEstimate } = require('openai-chat-tokens');
-const { EModelEndpoint, supportsBalanceCheck } = require('librechat-data-provider');
-const { formatFromLangChain } = require('~/app/clients/prompts');
-const { getBalanceConfig } = require('~/server/services/Config');
-const { checkBalance } = require('~/models/balanceMethods');
-const { logger } = require('~/config');
-
-const createStartHandler = ({
-  context,
-  conversationId,
-  tokenBuffer = 0,
-  initialMessageCount,
-  manager,
-}) => {
-  return async (_llm, _messages, runId, parentRunId, extraParams) => {
-    const { invocation_params } = extraParams;
-    const { model, functions, function_call } = invocation_params;
-    const messages = _messages[0].map(formatFromLangChain);
-
-    logger.debug(`[createStartHandler] handleChatModelStart: ${context}`, {
-      model,
-      function_call,
-    });
-
-    if (context !== 'title') {
-      logger.debug(`[createStartHandler] handleChatModelStart: ${context}`, {
-        functions,
-      });
-    }
-
-    const payload = { messages };
-    let prelimPromptTokens = 1;
-
-    if (functions) {
-      payload.functions = functions;
-      prelimPromptTokens += 2;
-    }
-
-    if (function_call) {
-      payload.function_call = function_call;
-      prelimPromptTokens -= 5;
-    }
-
-    prelimPromptTokens += promptTokensEstimate(payload);
-    logger.debug('[createStartHandler]', {
-      prelimPromptTokens,
-      tokenBuffer,
-    });
-    prelimPromptTokens += tokenBuffer;
-
-    try {
-      const balance = await getBalanceConfig();
-      if (balance?.enabled && supportsBalanceCheck[EModelEndpoint.openAI]) {
-        const generations =
-          initialMessageCount && messages.length > initialMessageCount
-            ? messages.slice(initialMessageCount)
-            : null;
-        await checkBalance({
-          req: manager.req,
-          res: manager.res,
-          txData: {
-            user: manager.user,
-            tokenType: 'prompt',
-            amount: prelimPromptTokens,
-            debug: manager.debug,
-            generations,
-            model,
-            endpoint: EModelEndpoint.openAI,
-          },
-        });
-      }
-    } catch (err) {
-      logger.error(`[createStartHandler][${context}] checkBalance error`, err);
-      manager.abortController.abort();
-      if (context === 'summary' || context === 'plugins') {
-        manager.addRun(runId, { conversationId, error: err.message });
-        throw new Error(err);
-      }
-      return;
-    }
-
-    manager.addRun(runId, {
-      model,
-      messages,
-      functions,
-      function_call,
-      runId,
-      parentRunId,
-      conversationId,
-      prelimPromptTokens,
-    });
-  };
-};
-
-module.exports = createStartHandler;

api/app/clients/callbacks/index.js

@@ -1,5 +0,0 @@
-const createStartHandler = require('./createStartHandler');
-
-module.exports = {
-  createStartHandler,
-};

api/app/clients/index.js

@@ -1,15 +1,11 @@
-const ChatGPTClient = require('./ChatGPTClient');
 const OpenAIClient = require('./OpenAIClient');
-const PluginsClient = require('./PluginsClient');
 const GoogleClient = require('./GoogleClient');
 const TextStream = require('./TextStream');
 const AnthropicClient = require('./AnthropicClient');
 const toolUtils = require('./tools/util');
 
 module.exports = {
-  ChatGPTClient,
   OpenAIClient,
-  PluginsClient,
   GoogleClient,
   TextStream,
   AnthropicClient,

api/app/clients/llm/RunManager.js

@@ -1,105 +0,0 @@
-const { createStartHandler } = require('~/app/clients/callbacks');
-const { spendTokens } = require('~/models/spendTokens');
-const { logger } = require('~/config');
-
-class RunManager {
-  constructor(fields) {
-    const { req, res, abortController, debug } = fields;
-    this.abortController = abortController;
-    this.user = req.user.id;
-    this.req = req;
-    this.res = res;
-    this.debug = debug;
-    this.runs = new Map();
-    this.convos = new Map();
-  }
-
-  addRun(runId, runData) {
-    if (!this.runs.has(runId)) {
-      this.runs.set(runId, runData);
-      if (runData.conversationId) {
-        this.convos.set(runData.conversationId, runId);
-      }
-      return runData;
-    } else {
-      const existingData = this.runs.get(runId);
-      const update = { ...existingData, ...runData };
-      this.runs.set(runId, update);
-      if (update.conversationId) {
-        this.convos.set(update.conversationId, runId);
-      }
-      return update;
-    }
-  }
-
-  removeRun(runId) {
-    if (this.runs.has(runId)) {
-      this.runs.delete(runId);
-    } else {
-      logger.error(`[api/app/clients/llm/RunManager] Run with ID ${runId} does not exist.`);
-    }
-  }
-
-  getAllRuns() {
-    return Array.from(this.runs.values());
-  }
-
-  getRunById(runId) {
-    return this.runs.get(runId);
-  }
-
-  getRunByConversationId(conversationId) {
-    const runId = this.convos.get(conversationId);
-    return { run: this.runs.get(runId), runId };
-  }
-
-  createCallbacks(metadata) {
-    return [
-      {
-        handleChatModelStart: createStartHandler({ ...metadata, manager: this }),
-        handleLLMEnd: async (output, runId, _parentRunId) => {
-          const { llmOutput, ..._output } = output;
-          logger.debug(`[RunManager] handleLLMEnd: ${JSON.stringify(metadata)}`, {
-            runId,
-            _parentRunId,
-            llmOutput,
-          });
-
-          if (metadata.context !== 'title') {
-            logger.debug('[RunManager] handleLLMEnd:', {
-              output: _output,
-            });
-          }
-
-          const { tokenUsage } = output.llmOutput;
-          const run = this.getRunById(runId);
-          this.removeRun(runId);
-
-          const txData = {
-            user: this.user,
-            model: run?.model ?? 'gpt-3.5-turbo',
-            ...metadata,
-          };
-
-          await spendTokens(txData, tokenUsage);
-        },
-        handleLLMError: async (err) => {
-          logger.error(`[RunManager] handleLLMError: ${JSON.stringify(metadata)}`, err);
-          if (metadata.context === 'title') {
-            return;
-          } else if (metadata.context === 'plugins') {
-            throw new Error(err);
-          }
-          const { conversationId } = metadata;
-          const { run } = this.getRunByConversationId(conversationId);
-          if (run && run.error) {
-            const { error } = run;
-            throw new Error(error);
-          }
-        },
-      },
-    ];
-  }
-}
-
-module.exports = RunManager;

api/app/clients/llm/index.js

@@ -1,9 +1,7 @@
 const createLLM = require('./createLLM');
-const RunManager = require('./RunManager');
 const createCoherePayload = require('./createCoherePayload');
 
 module.exports = {
   createLLM,
-  RunManager,
   createCoherePayload,
 };

api/app/clients/prompts/createContextHandlers.js

@@ -1,6 +1,6 @@
 const axios = require('axios');
-const { isEnabled } = require('~/server/utils');
-const { logger } = require('~/config');
+const { logger } = require('@librechat/data-schemas');
+const { isEnabled, generateShortLivedToken } = require('@librechat/api');
 
 const footer = `Use the context as your learned knowledge to better answer the user.
 
@@ -18,7 +18,7 @@ function createContextHandlers(req, userMessageContent) {
   const queryPromises = [];
   const processedFiles = [];
   const processedIds = new Set();
-  const jwtToken = req.headers.authorization.split(' ')[1];
+  const jwtToken = generateShortLivedToken(req.user.id);
   const useFullContext = isEnabled(process.env.RAG_USE_FULL_CONTEXT);
 
   const query = async (file) => {
@@ -96,35 +96,35 @@ function createContextHandlers(req, userMessageContent) {
         resolvedQueries.length === 0
           ? '\n\tThe semantic search did not return any results.'
           : resolvedQueries
-            .map((queryResult, index) => {
-              const file = processedFiles[index];
-              let contextItems = queryResult.data;
+              .map((queryResult, index) => {
+                const file = processedFiles[index];
+                let contextItems = queryResult.data;
 
-              const generateContext = (currentContext) =>
-                `
+                const generateContext = (currentContext) =>
+                  `
           <file>
             <filename>${file.filename}</filename>
             <context>${currentContext}
             </context>
           </file>`;
 
-              if (useFullContext) {
-                return generateContext(`\n${contextItems}`);
-              }
+                if (useFullContext) {
+                  return generateContext(`\n${contextItems}`);
+                }
 
-              contextItems = queryResult.data
-                .map((item) => {
-                  const pageContent = item[0].page_content;
-                  return `
+                contextItems = queryResult.data
+                  .map((item) => {
+                    const pageContent = item[0].page_content;
+                    return `
             <contextItem>
               <![CDATA[${pageContent?.trim()}]]>
             </contextItem>`;
-                })
-                .join('');
+                  })
+                  .join('');
 
-              return generateContext(contextItems);
-            })
-            .join('');
+                return generateContext(contextItems);
+              })
+              .join('');
 
       if (useFullContext) {
         const prompt = `${header}

api/app/clients/prompts/formatMessages.js

@@ -237,41 +237,9 @@ const formatAgentMessages = (payload) => {
   return messages;
 };
 
-/**
- * Formats an array of messages for LangChain, making sure all content fields are strings
- * @param {Array<(HumanMessage|AIMessage|SystemMessage|ToolMessage)>} payload - The array of messages to format.
- * @returns {Array<(HumanMessage|AIMessage|SystemMessage|ToolMessage)>} - The array of formatted LangChain messages, including ToolMessages for tool calls.
- */
-const formatContentStrings = (payload) => {
-  const messages = [];
-
-  for (const message of payload) {
-    if (typeof message.content === 'string') {
-      continue;
-    }
-
-    if (!Array.isArray(message.content)) {
-      continue;
-    }
-
-    // Reduce text types to a single string, ignore all other types
-    const content = message.content.reduce((acc, curr) => {
-      if (curr.type === ContentTypes.TEXT) {
-        return `${acc}${curr[ContentTypes.TEXT]}\n`;
-      }
-      return acc;
-    }, '');
-
-    message.content = content.trim();
-  }
-
-  return messages;
-};
-
 module.exports = {
   formatMessage,
   formatFromLangChain,
   formatAgentMessages,
-  formatContentStrings,
   formatLangChainMessages,
 };

api/app/clients/specs/AnthropicClient.test.js

@@ -245,7 +245,7 @@ describe('AnthropicClient', () => {
     });
 
     describe('Claude 4 model headers', () => {
-      it('should add "prompt-caching" beta header for claude-sonnet-4 model', () => {
+      it('should add "prompt-caching" and "context-1m" beta headers for claude-sonnet-4 model', () => {
         const client = new AnthropicClient('test-api-key');
         const modelOptions = {
           model: 'claude-sonnet-4-20250514',
@@ -255,10 +255,30 @@ describe('AnthropicClient', () => {
         expect(anthropicClient._options.defaultHeaders).toBeDefined();
         expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
         expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
-          'prompt-caching-2024-07-31',
+          'prompt-caching-2024-07-31,context-1m-2025-08-07',
         );
       });
 
+      it('should add "prompt-caching" and "context-1m" beta headers for claude-sonnet-4 model formats', () => {
+        const client = new AnthropicClient('test-api-key');
+        const modelVariations = [
+          'claude-sonnet-4-20250514',
+          'claude-sonnet-4-latest',
+          'anthropic/claude-sonnet-4-20250514',
+        ];
+
+        modelVariations.forEach((model) => {
+          const modelOptions = { model };
+          client.setOptions({ modelOptions, promptCache: true });
+          const anthropicClient = client.getClient(modelOptions);
+          expect(anthropicClient._options.defaultHeaders).toBeDefined();
+          expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
+          expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
+            'prompt-caching-2024-07-31,context-1m-2025-08-07',
+          );
+        });
+      });
+
       it('should add "prompt-caching" beta header for claude-opus-4 model', () => {
         const client = new AnthropicClient('test-api-key');
         const modelOptions = {
@@ -273,20 +293,6 @@ describe('AnthropicClient', () => {
         );
       });
 
-      it('should add "prompt-caching" beta header for claude-4-sonnet model', () => {
-        const client = new AnthropicClient('test-api-key');
-        const modelOptions = {
-          model: 'claude-4-sonnet-20250514',
-        };
-        client.setOptions({ modelOptions, promptCache: true });
-        const anthropicClient = client.getClient(modelOptions);
-        expect(anthropicClient._options.defaultHeaders).toBeDefined();
-        expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
-        expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
-          'prompt-caching-2024-07-31',
-        );
-      });
-
       it('should add "prompt-caching" beta header for claude-4-opus model', () => {
         const client = new AnthropicClient('test-api-key');
         const modelOptions = {
@@ -309,7 +315,7 @@ describe('AnthropicClient', () => {
       };
       client.setOptions({ modelOptions, promptCache: true });
       const anthropicClient = client.getClient(modelOptions);
-      expect(anthropicClient.defaultHeaders).not.toHaveProperty('anthropic-beta');
+      expect(anthropicClient._options.defaultHeaders).toBeUndefined();
     });
 
     it('should not add beta header for other models', () => {
@@ -320,7 +326,7 @@ describe('AnthropicClient', () => {
         },
       });
       const anthropicClient = client.getClient();
-      expect(anthropicClient.defaultHeaders).not.toHaveProperty('anthropic-beta');
+      expect(anthropicClient._options.defaultHeaders).toBeUndefined();
     });
   });
 

api/app/clients/specs/BaseClient.test.js

@@ -2,6 +2,14 @@ const { Constants } = require('librechat-data-provider');
 const { initializeFakeClient } = require('./FakeClient');
 
 jest.mock('~/db/connect');
+jest.mock('~/server/services/Config', () => ({
+  getAppConfig: jest.fn().mockResolvedValue({
+    // Default app config for tests
+    paths: { uploads: '/tmp' },
+    fileStrategy: 'local',
+    memory: { disabled: false },
+  }),
+}));
 jest.mock('~/models', () => ({
   User: jest.fn(),
   Key: jest.fn(),
@@ -422,6 +430,46 @@ describe('BaseClient', () => {
       expect(response).toEqual(expectedResult);
     });
 
+    test('should replace responseMessageId with new UUID when isRegenerate is true and messageId ends with underscore', async () => {
+      const mockCrypto = require('crypto');
+      const newUUID = 'new-uuid-1234';
+      jest.spyOn(mockCrypto, 'randomUUID').mockReturnValue(newUUID);
+
+      const opts = {
+        isRegenerate: true,
+        responseMessageId: 'existing-message-id_',
+      };
+
+      await TestClient.setMessageOptions(opts);
+
+      expect(TestClient.responseMessageId).toBe(newUUID);
+      expect(TestClient.responseMessageId).not.toBe('existing-message-id_');
+
+      mockCrypto.randomUUID.mockRestore();
+    });
+
+    test('should not replace responseMessageId when isRegenerate is false', async () => {
+      const opts = {
+        isRegenerate: false,
+        responseMessageId: 'existing-message-id_',
+      };
+
+      await TestClient.setMessageOptions(opts);
+
+      expect(TestClient.responseMessageId).toBe('existing-message-id_');
+    });
+
+    test('should not replace responseMessageId when it does not end with underscore', async () => {
+      const opts = {
+        isRegenerate: true,
+        responseMessageId: 'existing-message-id',
+      };
+
+      await TestClient.setMessageOptions(opts);
+
+      expect(TestClient.responseMessageId).toBe('existing-message-id');
+    });
+
     test('sendMessage should work with provided conversationId and parentMessageId', async () => {
       const userMessage = 'Second message in the conversation';
       const opts = {
@@ -539,6 +587,8 @@ describe('BaseClient', () => {
       expect(onStart).toHaveBeenCalledWith(
         expect.objectContaining({ text: 'Hello, world!' }),
         expect.any(String),
+        /** `isNewConvo` */
+        true,
       );
     });
 

api/app/clients/specs/FakeClient.js

@@ -1,5 +1,5 @@
+const { getModelMaxTokens } = require('@librechat/api');
 const BaseClient = require('../BaseClient');
-const { getModelMaxTokens } = require('../../../utils');
 
 class FakeClient extends BaseClient {
   constructor(apiKey, options = {}) {

api/app/clients/specs/OpenAIClient.test.js

@@ -531,44 +531,6 @@ describe('OpenAIClient', () => {
     });
   });
 
-  describe('sendMessage/getCompletion/chatCompletion', () => {
-    afterEach(() => {
-      delete process.env.AZURE_OPENAI_DEFAULT_MODEL;
-      delete process.env.AZURE_USE_MODEL_AS_DEPLOYMENT_NAME;
-    });
-
-    it('should call getCompletion and fetchEventSource when using a text/instruct model', async () => {
-      const model = 'text-davinci-003';
-      const onProgress = jest.fn().mockImplementation(() => ({}));
-
-      const testClient = new OpenAIClient('test-api-key', {
-        ...defaultOptions,
-        modelOptions: { model },
-      });
-
-      const getCompletion = jest.spyOn(testClient, 'getCompletion');
-      await testClient.sendMessage('Hi mom!', { onProgress });
-
-      expect(getCompletion).toHaveBeenCalled();
-      expect(getCompletion.mock.calls.length).toBe(1);
-
-      expect(getCompletion.mock.calls[0][0]).toBe('||>User:\nHi mom!\n||>Assistant:\n');
-
-      expect(fetchEventSource).toHaveBeenCalled();
-      expect(fetchEventSource.mock.calls.length).toBe(1);
-
-      // Check if the first argument (url) is correct
-      const firstCallArgs = fetchEventSource.mock.calls[0];
-
-      const expectedURL = 'https://api.openai.com/v1/completions';
-      expect(firstCallArgs[0]).toBe(expectedURL);
-
-      const requestBody = JSON.parse(firstCallArgs[1].body);
-      expect(requestBody).toHaveProperty('model');
-      expect(requestBody.model).toBe(model);
-    });
-  });
-
   describe('checkVisionRequest functionality', () => {
     let client;
     const attachments = [{ type: 'image/png' }];

api/app/clients/specs/PluginsClient.test.js

@@ -1,314 +0,0 @@
-const crypto = require('crypto');
-const { Constants } = require('librechat-data-provider');
-const { HumanMessage, AIMessage } = require('@langchain/core/messages');
-const PluginsClient = require('../PluginsClient');
-
-jest.mock('~/db/connect');
-jest.mock('~/models/Conversation', () => {
-  return function () {
-    return {
-      save: jest.fn(),
-      deleteConvos: jest.fn(),
-    };
-  };
-});
-
-const defaultAzureOptions = {
-  azureOpenAIApiInstanceName: 'your-instance-name',
-  azureOpenAIApiDeploymentName: 'your-deployment-name',
-  azureOpenAIApiVersion: '2020-07-01-preview',
-};
-
-describe('PluginsClient', () => {
-  let TestAgent;
-  let options = {
-    tools: [],
-    modelOptions: {
-      model: 'gpt-3.5-turbo',
-      temperature: 0,
-      max_tokens: 2,
-    },
-    agentOptions: {
-      model: 'gpt-3.5-turbo',
-    },
-  };
-  let parentMessageId;
-  let conversationId;
-  const fakeMessages = [];
-  const userMessage = 'Hello, ChatGPT!';
-  const apiKey = 'fake-api-key';
-
-  beforeEach(() => {
-    TestAgent = new PluginsClient(apiKey, options);
-    TestAgent.loadHistory = jest
-      .fn()
-      .mockImplementation((conversationId, parentMessageId = null) => {
-        if (!conversationId) {
-          TestAgent.currentMessages = [];
-          return Promise.resolve([]);
-        }
-
-        const orderedMessages = TestAgent.constructor.getMessagesForConversation({
-          messages: fakeMessages,
-          parentMessageId,
-        });
-
-        const chatMessages = orderedMessages.map((msg) =>
-          msg?.isCreatedByUser || msg?.role?.toLowerCase() === 'user'
-            ? new HumanMessage(msg.text)
-            : new AIMessage(msg.text),
-        );
-
-        TestAgent.currentMessages = orderedMessages;
-        return Promise.resolve(chatMessages);
-      });
-    TestAgent.sendMessage = jest.fn().mockImplementation(async (message, opts = {}) => {
-      if (opts && typeof opts === 'object') {
-        TestAgent.setOptions(opts);
-      }
-      const conversationId = opts.conversationId || crypto.randomUUID();
-      const parentMessageId = opts.parentMessageId || Constants.NO_PARENT;
-      const userMessageId = opts.overrideParentMessageId || crypto.randomUUID();
-      this.pastMessages = await TestAgent.loadHistory(
-        conversationId,
-        TestAgent.options?.parentMessageId,
-      );
-
-      const userMessage = {
-        text: message,
-        sender: 'ChatGPT',
-        isCreatedByUser: true,
-        messageId: userMessageId,
-        parentMessageId,
-        conversationId,
-      };
-
-      const response = {
-        sender: 'ChatGPT',
-        text: 'Hello, User!',
-        isCreatedByUser: false,
-        messageId: crypto.randomUUID(),
-        parentMessageId: userMessage.messageId,
-        conversationId,
-      };
-
-      fakeMessages.push(userMessage);
-      fakeMessages.push(response);
-      return response;
-    });
-  });
-
-  test('initializes PluginsClient without crashing', () => {
-    expect(TestAgent).toBeInstanceOf(PluginsClient);
-  });
-
-  test('check setOptions function', () => {
-    expect(TestAgent.agentIsGpt3).toBe(true);
-  });
-
-  describe('sendMessage', () => {
-    test('sendMessage should return a response message', async () => {
-      const expectedResult = expect.objectContaining({
-        sender: 'ChatGPT',
-        text: expect.any(String),
-        isCreatedByUser: false,
-        messageId: expect.any(String),
-        parentMessageId: expect.any(String),
-        conversationId: expect.any(String),
-      });
-
-      const response = await TestAgent.sendMessage(userMessage);
-      parentMessageId = response.messageId;
-      conversationId = response.conversationId;
-      expect(response).toEqual(expectedResult);
-    });
-
-    test('sendMessage should work with provided conversationId and parentMessageId', async () => {
-      const userMessage = 'Second message in the conversation';
-      const opts = {
-        conversationId,
-        parentMessageId,
-      };
-
-      const expectedResult = expect.objectContaining({
-        sender: 'ChatGPT',
-        text: expect.any(String),
-        isCreatedByUser: false,
-        messageId: expect.any(String),
-        parentMessageId: expect.any(String),
-        conversationId: opts.conversationId,
-      });
-
-      const response = await TestAgent.sendMessage(userMessage, opts);
-      parentMessageId = response.messageId;
-      expect(response.conversationId).toEqual(conversationId);
-      expect(response).toEqual(expectedResult);
-    });
-
-    test('should return chat history', async () => {
-      const chatMessages = await TestAgent.loadHistory(conversationId, parentMessageId);
-      expect(TestAgent.currentMessages).toHaveLength(4);
-      expect(chatMessages[0].text).toEqual(userMessage);
-    });
-  });
-
-  describe('getFunctionModelName', () => {
-    let client;
-
-    beforeEach(() => {
-      client = new PluginsClient('dummy_api_key');
-    });
-
-    test('should return the input when it includes a dash followed by four digits', () => {
-      expect(client.getFunctionModelName('-1234')).toBe('-1234');
-      expect(client.getFunctionModelName('gpt-4-5678-preview')).toBe('gpt-4-5678-preview');
-    });
-
-    test('should return the input for all function-capable models (`0613` models and above)', () => {
-      expect(client.getFunctionModelName('gpt-4-0613')).toBe('gpt-4-0613');
-      expect(client.getFunctionModelName('gpt-4-32k-0613')).toBe('gpt-4-32k-0613');
-      expect(client.getFunctionModelName('gpt-3.5-turbo-0613')).toBe('gpt-3.5-turbo-0613');
-      expect(client.getFunctionModelName('gpt-3.5-turbo-16k-0613')).toBe('gpt-3.5-turbo-16k-0613');
-      expect(client.getFunctionModelName('gpt-3.5-turbo-1106')).toBe('gpt-3.5-turbo-1106');
-      expect(client.getFunctionModelName('gpt-4-1106-preview')).toBe('gpt-4-1106-preview');
-      expect(client.getFunctionModelName('gpt-4-1106')).toBe('gpt-4-1106');
-    });
-
-    test('should return the corresponding model if input is non-function capable (`0314` models)', () => {
-      expect(client.getFunctionModelName('gpt-4-0314')).toBe('gpt-4');
-      expect(client.getFunctionModelName('gpt-4-32k-0314')).toBe('gpt-4');
-      expect(client.getFunctionModelName('gpt-3.5-turbo-0314')).toBe('gpt-3.5-turbo');
-      expect(client.getFunctionModelName('gpt-3.5-turbo-16k-0314')).toBe('gpt-3.5-turbo');
-    });
-
-    test('should return "gpt-3.5-turbo" when the input includes "gpt-3.5-turbo"', () => {
-      expect(client.getFunctionModelName('test gpt-3.5-turbo model')).toBe('gpt-3.5-turbo');
-    });
-
-    test('should return "gpt-4" when the input includes "gpt-4"', () => {
-      expect(client.getFunctionModelName('testing gpt-4')).toBe('gpt-4');
-    });
-
-    test('should return "gpt-3.5-turbo" for input that does not meet any specific condition', () => {
-      expect(client.getFunctionModelName('random string')).toBe('gpt-3.5-turbo');
-      expect(client.getFunctionModelName('')).toBe('gpt-3.5-turbo');
-    });
-  });
-
-  describe('Azure OpenAI tests specific to Plugins', () => {
-    // TODO: add more tests for Azure OpenAI integration with Plugins
-    // let client;
-    // beforeEach(() => {
-    //   client = new PluginsClient('dummy_api_key');
-    // });
-
-    test('should not call getFunctionModelName when azure options are set', () => {
-      const spy = jest.spyOn(PluginsClient.prototype, 'getFunctionModelName');
-      const model = 'gpt-4-turbo';
-
-      // note, without the azure change in PR #1766, `getFunctionModelName` is called twice
-      const testClient = new PluginsClient('dummy_api_key', {
-        agentOptions: {
-          model,
-          agent: 'functions',
-        },
-        azure: defaultAzureOptions,
-      });
-
-      expect(spy).not.toHaveBeenCalled();
-      expect(testClient.agentOptions.model).toBe(model);
-
-      spy.mockRestore();
-    });
-  });
-
-  describe('sendMessage with filtered tools', () => {
-    let TestAgent;
-    const apiKey = 'fake-api-key';
-    const mockTools = [{ name: 'tool1' }, { name: 'tool2' }, { name: 'tool3' }, { name: 'tool4' }];
-
-    beforeEach(() => {
-      TestAgent = new PluginsClient(apiKey, {
-        tools: mockTools,
-        modelOptions: {
-          model: 'gpt-3.5-turbo',
-          temperature: 0,
-          max_tokens: 2,
-        },
-        agentOptions: {
-          model: 'gpt-3.5-turbo',
-        },
-      });
-
-      TestAgent.options.req = {
-        app: {
-          locals: {},
-        },
-      };
-
-      TestAgent.sendMessage = jest.fn().mockImplementation(async () => {
-        const { filteredTools = [], includedTools = [] } = TestAgent.options.req.app.locals;
-
-        if (includedTools.length > 0) {
-          const tools = TestAgent.options.tools.filter((plugin) =>
-            includedTools.includes(plugin.name),
-          );
-          TestAgent.options.tools = tools;
-        } else {
-          const tools = TestAgent.options.tools.filter(
-            (plugin) => !filteredTools.includes(plugin.name),
-          );
-          TestAgent.options.tools = tools;
-        }
-
-        return {
-          text: 'Mocked response',
-          tools: TestAgent.options.tools,
-        };
-      });
-    });
-
-    test('should filter out tools when filteredTools is provided', async () => {
-      TestAgent.options.req.app.locals.filteredTools = ['tool1', 'tool3'];
-      const response = await TestAgent.sendMessage('Test message');
-      expect(response.tools).toHaveLength(2);
-      expect(response.tools).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({ name: 'tool2' }),
-          expect.objectContaining({ name: 'tool4' }),
-        ]),
-      );
-    });
-
-    test('should only include specified tools when includedTools is provided', async () => {
-      TestAgent.options.req.app.locals.includedTools = ['tool2', 'tool4'];
-      const response = await TestAgent.sendMessage('Test message');
-      expect(response.tools).toHaveLength(2);
-      expect(response.tools).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({ name: 'tool2' }),
-          expect.objectContaining({ name: 'tool4' }),
-        ]),
-      );
-    });
-
-    test('should prioritize includedTools over filteredTools', async () => {
-      TestAgent.options.req.app.locals.filteredTools = ['tool1', 'tool3'];
-      TestAgent.options.req.app.locals.includedTools = ['tool1', 'tool2'];
-      const response = await TestAgent.sendMessage('Test message');
-      expect(response.tools).toHaveLength(2);
-      expect(response.tools).toEqual(
-        expect.arrayContaining([
-          expect.objectContaining({ name: 'tool1' }),
-          expect.objectContaining({ name: 'tool2' }),
-        ]),
-      );
-    });
-
-    test('should not modify tools when no filters are provided', async () => {
-      const response = await TestAgent.sendMessage('Test message');
-      expect(response.tools).toHaveLength(4);
-      expect(response.tools).toEqual(expect.arrayContaining(mockTools));
-    });
-  });
-});

api/app/clients/tools/index.js

@@ -1,4 +1,4 @@
-const availableTools = require('./manifest.json');
+const manifest = require('./manifest');
 
 // Structured Tools
 const DALLE3 = require('./structured/DALLE3');
@@ -13,23 +13,8 @@ const TraversaalSearch = require('./structured/TraversaalSearch');
 const createOpenAIImageTools = require('./structured/OpenAIImageTools');
 const TavilySearchResults = require('./structured/TavilySearchResults');
 
-/** @type {Record<string, TPlugin | undefined>} */
-const manifestToolMap = {};
-
-/** @type {Array<TPlugin>} */
-const toolkits = [];
-
-availableTools.forEach((tool) => {
-  manifestToolMap[tool.pluginKey] = tool;
-  if (tool.toolkit === true) {
-    toolkits.push(tool);
-  }
-});
-
 module.exports = {
-  toolkits,
-  availableTools,
-  manifestToolMap,
+  ...manifest,
   // Structured Tools
   DALLE3,
   FluxAPI,

api/app/clients/tools/manifest.js

@@ -0,0 +1,20 @@
+const availableTools = require('./manifest.json');
+
+/** @type {Record<string, TPlugin | undefined>} */
+const manifestToolMap = {};
+
+/** @type {Array<TPlugin>} */
+const toolkits = [];
+
+availableTools.forEach((tool) => {
+  manifestToolMap[tool.pluginKey] = tool;
+  if (tool.toolkit === true) {
+    toolkits.push(tool);
+  }
+});
+
+module.exports = {
+  toolkits,
+  availableTools,
+  manifestToolMap,
+};

api/app/clients/tools/manifest.json

@@ -49,7 +49,7 @@
     "pluginKey": "image_gen_oai",
     "toolkit": true,
     "description": "Image Generation and Editing using OpenAI's latest state-of-the-art models",
-    "icon": "/assets/image_gen_oai.png",
+    "icon": "assets/image_gen_oai.png",
     "authConfig": [
       {
         "authField": "IMAGE_GEN_OAI_API_KEY",
@@ -75,7 +75,7 @@
     "name": "Browser",
     "pluginKey": "web-browser",
     "description": "Scrape and summarize webpage data",
-    "icon": "/assets/web-browser.svg",
+    "icon": "assets/web-browser.svg",
     "authConfig": [
       {
         "authField": "OPENAI_API_KEY",
@@ -170,7 +170,7 @@
     "name": "OpenWeather",
     "pluginKey": "open_weather",
     "description": "Get weather forecasts and historical data from the OpenWeather API",
-    "icon": "/assets/openweather.png",
+    "icon": "assets/openweather.png",
     "authConfig": [
       {
         "authField": "OPENWEATHER_API_KEY",

api/app/clients/tools/structured/DALLE3.js

@@ -3,12 +3,12 @@ const path = require('path');
 const OpenAI = require('openai');
 const fetch = require('node-fetch');
 const { v4: uuidv4 } = require('uuid');
+const { ProxyAgent } = require('undici');
 const { Tool } = require('@langchain/core/tools');
-const { HttpsProxyAgent } = require('https-proxy-agent');
+const { logger } = require('@librechat/data-schemas');
+const { getImageBasename } = require('@librechat/api');
 const { FileContext, ContentTypes } = require('librechat-data-provider');
-const { getImageBasename } = require('~/server/services/Files/images');
 const extractBaseURL = require('~/utils/extractBaseURL');
-const logger = require('~/config/winston');
 
 const displayMessage =
   "DALL-E displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
@@ -46,7 +46,10 @@ class DALLE3 extends Tool {
     }
 
     if (process.env.PROXY) {
-      config.httpAgent = new HttpsProxyAgent(process.env.PROXY);
+      const proxyAgent = new ProxyAgent(process.env.PROXY);
+      config.fetchOptions = {
+        dispatcher: proxyAgent,
+      };
     }
 
     /** @type {OpenAI} */
@@ -163,7 +166,8 @@ Error Message: ${error.message}`);
     if (this.isAgent) {
       let fetchOptions = {};
       if (process.env.PROXY) {
-        fetchOptions.agent = new HttpsProxyAgent(process.env.PROXY);
+        const proxyAgent = new ProxyAgent(process.env.PROXY);
+        fetchOptions.dispatcher = proxyAgent;
       }
       const imageResponse = await fetch(theImageUrl, fetchOptions);
       const arrayBuffer = await imageResponse.arrayBuffer();

api/app/clients/tools/structured/OpenAIImageTools.js

@@ -1,69 +1,16 @@
-const { z } = require('zod');
 const axios = require('axios');
 const { v4 } = require('uuid');
 const OpenAI = require('openai');
 const FormData = require('form-data');
+const { ProxyAgent } = require('undici');
 const { tool } = require('@langchain/core/tools');
-const { logAxiosError } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
-const { HttpsProxyAgent } = require('https-proxy-agent');
+const { logAxiosError, oaiToolkit } = require('@librechat/api');
 const { ContentTypes, EImageOutputType } = require('librechat-data-provider');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
-const { extractBaseURL } = require('~/utils');
+const extractBaseURL = require('~/utils/extractBaseURL');
 const { getFiles } = require('~/models/File');
 
-/** Default descriptions for image generation tool  */
-const DEFAULT_IMAGE_GEN_DESCRIPTION = `
-Generates high-quality, original images based solely on text, not using any uploaded reference images.
-
-When to use \`image_gen_oai\`:
-- To create entirely new images from detailed text descriptions that do NOT reference any image files.
-
-When NOT to use \`image_gen_oai\`:
-- If the user has uploaded any images and requests modifications, enhancements, or remixing based on those uploads → use \`image_edit_oai\` instead.
-
-Generated image IDs will be returned in the response, so you can refer to them in future requests made to \`image_edit_oai\`.
-`.trim();
-
-/** Default description for image editing tool  */
-const DEFAULT_IMAGE_EDIT_DESCRIPTION =
-  `Generates high-quality, original images based on text and one or more uploaded/referenced images.
-
-When to use \`image_edit_oai\`:
-- The user wants to modify, extend, or remix one **or more** uploaded images, either:
-- Previously generated, or in the current request (both to be included in the \`image_ids\` array).
-- Always when the user refers to uploaded images for editing, enhancement, remixing, style transfer, or combining elements.
-- Any current or existing images are to be used as visual guides.
-- If there are any files in the current request, they are more likely than not expected as references for image edit requests.
-
-When NOT to use \`image_edit_oai\`:
-- Brand-new generations that do not rely on an existing image → use \`image_gen_oai\` instead.
-
-Both generated and referenced image IDs will be returned in the response, so you can refer to them in future requests made to \`image_edit_oai\`.
-`.trim();
-
-/** Default prompt descriptions  */
-const DEFAULT_IMAGE_GEN_PROMPT_DESCRIPTION = `Describe the image you want in detail. 
-      Be highly specific—break your idea into layers: 
-      (1) main concept and subject,
-      (2) composition and position,
-      (3) lighting and mood,
-      (4) style, medium, or camera details,
-      (5) important features (age, expression, clothing, etc.),
-      (6) background.
-      Use positive, descriptive language and specify what should be included, not what to avoid. 
-      List number and characteristics of people/objects, and mention style/technical requirements (e.g., "DSLR photo, 85mm lens, golden hour").
-      Do not reference any uploaded images—use for new image creation from text only.`;
-
-const DEFAULT_IMAGE_EDIT_PROMPT_DESCRIPTION = `Describe the changes, enhancements, or new ideas to apply to the uploaded image(s).
-      Be highly specific—break your request into layers: 
-      (1) main concept or transformation,
-      (2) specific edits/replacements or composition guidance,
-      (3) desired style, mood, or technique,
-      (4) features/items to keep, change, or add (such as objects, people, clothing, lighting, etc.).
-      Use positive, descriptive language and clarify what should be included or changed, not what to avoid.
-      Always base this prompt on the most recently uploaded reference images.`;
-
 const displayMessage =
   "The tool displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
 
@@ -91,21 +38,11 @@ function returnValue(value) {
   return value;
 }
 
-const getImageGenDescription = () => {
-  return process.env.IMAGE_GEN_OAI_DESCRIPTION || DEFAULT_IMAGE_GEN_DESCRIPTION;
-};
-
-const getImageEditDescription = () => {
-  return process.env.IMAGE_EDIT_OAI_DESCRIPTION || DEFAULT_IMAGE_EDIT_DESCRIPTION;
-};
-
-const getImageGenPromptDescription = () => {
-  return process.env.IMAGE_GEN_OAI_PROMPT_DESCRIPTION || DEFAULT_IMAGE_GEN_PROMPT_DESCRIPTION;
-};
-
-const getImageEditPromptDescription = () => {
-  return process.env.IMAGE_EDIT_OAI_PROMPT_DESCRIPTION || DEFAULT_IMAGE_EDIT_PROMPT_DESCRIPTION;
-};
+function createAbortHandler() {
+  return function () {
+    logger.debug('[ImageGenOAI] Image generation aborted');
+  };
+}
 
 /**
  * Creates OpenAI Image tools (generation and editing)
@@ -115,7 +52,9 @@ const getImageEditPromptDescription = () => {
  * @param {string} fields.IMAGE_GEN_OAI_API_KEY - The OpenAI API key
  * @param {boolean} [fields.override] - Whether to override the API key check, necessary for app initialization
  * @param {MongoFile[]} [fields.imageFiles] - The images to be used for editing
- * @returns {Array} - Array of image tools
+ * @param {string} [fields.imageOutputType] - The image output type configuration
+ * @param {string} [fields.fileStrategy] - The file storage strategy
+ * @returns {Array<ReturnType<tool>>} - Array of image tools
  */
 function createOpenAIImageTools(fields = {}) {
   /** @type {boolean} Used to initialize the Tool without necessary variables. */
@@ -125,8 +64,8 @@ function createOpenAIImageTools(fields = {}) {
     throw new Error('This tool is only available for agents.');
   }
   const { req } = fields;
-  const imageOutputType = req?.app.locals.imageOutputType || EImageOutputType.PNG;
-  const appFileStrategy = req?.app.locals.fileStrategy;
+  const imageOutputType = fields.imageOutputType || EImageOutputType.PNG;
+  const appFileStrategy = fields.fileStrategy;
 
   const getApiKey = () => {
     const apiKey = process.env.IMAGE_GEN_OAI_API_KEY ?? '';
@@ -183,7 +122,10 @@ function createOpenAIImageTools(fields = {}) {
       }
       const clientConfig = { ...closureConfig };
       if (process.env.PROXY) {
-        clientConfig.httpAgent = new HttpsProxyAgent(process.env.PROXY);
+        const proxyAgent = new ProxyAgent(process.env.PROXY);
+        clientConfig.fetchOptions = {
+          dispatcher: proxyAgent,
+        };
       }
 
       /** @type {OpenAI} */
@@ -201,10 +143,18 @@ function createOpenAIImageTools(fields = {}) {
       }
 
       let resp;
+      /** @type {AbortSignal} */
+      let derivedSignal = null;
+      /** @type {() => void} */
+      let abortHandler = null;
+
       try {
-        const derivedSignal = runnableConfig?.signal
-          ? AbortSignal.any([runnableConfig.signal])
-          : undefined;
+        if (runnableConfig?.signal) {
+          derivedSignal = AbortSignal.any([runnableConfig.signal]);
+          abortHandler = createAbortHandler();
+          derivedSignal.addEventListener('abort', abortHandler, { once: true });
+        }
+
         resp = await openai.images.generate(
           {
             model: 'gpt-image-1',
@@ -228,6 +178,10 @@ function createOpenAIImageTools(fields = {}) {
         logAxiosError({ error, message });
         return returnValue(`Something went wrong when trying to generate the image. The OpenAI API may be unavailable:
 Error Message: ${error.message}`);
+      } finally {
+        if (abortHandler && derivedSignal) {
+          derivedSignal.removeEventListener('abort', abortHandler);
+        }
       }
 
       if (!resp) {
@@ -264,46 +218,7 @@ Error Message: ${error.message}`);
       ];
       return [response, { content, file_ids }];
     },
-    {
-      name: 'image_gen_oai',
-      description: getImageGenDescription(),
-      schema: z.object({
-        prompt: z.string().max(32000).describe(getImageGenPromptDescription()),
-        background: z
-          .enum(['transparent', 'opaque', 'auto'])
-          .optional()
-          .describe(
-            'Sets transparency for the background. Must be one of transparent, opaque or auto (default). When transparent, the output format should be png or webp.',
-          ),
-        /*
-        n: z
-          .number()
-          .int()
-          .min(1)
-          .max(10)
-          .optional()
-          .describe('The number of images to generate. Must be between 1 and 10.'),
-        output_compression: z
-          .number()
-          .int()
-          .min(0)
-          .max(100)
-          .optional()
-          .describe('The compression level (0-100%) for webp or jpeg formats. Defaults to 100.'),
-           */
-        quality: z
-          .enum(['auto', 'high', 'medium', 'low'])
-          .optional()
-          .describe('The quality of the image. One of auto (default), high, medium, or low.'),
-        size: z
-          .enum(['auto', '1024x1024', '1536x1024', '1024x1536'])
-          .optional()
-          .describe(
-            'The size of the generated image. One of 1024x1024, 1536x1024 (landscape), 1024x1536 (portrait), or auto (default).',
-          ),
-      }),
-      responseFormat: 'content_and_artifact',
-    },
+    oaiToolkit.image_gen_oai,
   );
 
   /**
@@ -317,7 +232,10 @@ Error Message: ${error.message}`);
 
       const clientConfig = { ...closureConfig };
       if (process.env.PROXY) {
-        clientConfig.httpAgent = new HttpsProxyAgent(process.env.PROXY);
+        const proxyAgent = new ProxyAgent(process.env.PROXY);
+        clientConfig.fetchOptions = {
+          dispatcher: proxyAgent,
+        };
       }
 
       const formData = new FormData();
@@ -409,10 +327,17 @@ Error Message: ${error.message}`);
         headers['Authorization'] = `Bearer ${apiKey}`;
       }
 
+      /** @type {AbortSignal} */
+      let derivedSignal = null;
+      /** @type {() => void} */
+      let abortHandler = null;
+
       try {
-        const derivedSignal = runnableConfig?.signal
-          ? AbortSignal.any([runnableConfig.signal])
-          : undefined;
+        if (runnableConfig?.signal) {
+          derivedSignal = AbortSignal.any([runnableConfig.signal]);
+          abortHandler = createAbortHandler();
+          derivedSignal.addEventListener('abort', abortHandler, { once: true });
+        }
 
         /** @type {import('axios').AxiosRequestConfig} */
         const axiosConfig = {
@@ -422,6 +347,19 @@ Error Message: ${error.message}`);
           baseURL,
         };
 
+        if (process.env.PROXY) {
+          try {
+            const url = new URL(process.env.PROXY);
+            axiosConfig.proxy = {
+              host: url.hostname.replace(/^\[|\]$/g, ''),
+              port: url.port ? parseInt(url.port, 10) : undefined,
+              protocol: url.protocol.replace(':', ''),
+            };
+          } catch (error) {
+            logger.error('Error parsing proxy URL:', error);
+          }
+        }
+
         if (process.env.IMAGE_GEN_OAI_AZURE_API_VERSION && process.env.IMAGE_GEN_OAI_BASEURL) {
           axiosConfig.params = {
             'api-version': process.env.IMAGE_GEN_OAI_AZURE_API_VERSION,
@@ -467,50 +405,13 @@ Error Message: ${error.message}`);
         logAxiosError({ error, message });
         return returnValue(`Something went wrong when trying to edit the image. The OpenAI API may be unavailable:
 Error Message: ${error.message || 'Unknown error'}`);
+      } finally {
+        if (abortHandler && derivedSignal) {
+          derivedSignal.removeEventListener('abort', abortHandler);
+        }
       }
     },
-    {
-      name: 'image_edit_oai',
-      description: getImageEditDescription(),
-      schema: z.object({
-        image_ids: z
-          .array(z.string())
-          .min(1)
-          .describe(
-            `
-IDs (image ID strings) of previously generated or uploaded images that should guide the edit.
-
-Guidelines:
-- If the user's request depends on any prior image(s), copy their image IDs into the \`image_ids\` array (in the same order the user refers to them).  
-- Never invent or hallucinate IDs; only use IDs that are still visible in the conversation context.
-- If no earlier image is relevant, omit the field entirely.
-`.trim(),
-          ),
-        prompt: z.string().max(32000).describe(getImageEditPromptDescription()),
-        /*
-        n: z
-          .number()
-          .int()
-          .min(1)
-          .max(10)
-          .optional()
-          .describe('The number of images to generate. Must be between 1 and 10. Defaults to 1.'),
-        */
-        quality: z
-          .enum(['auto', 'high', 'medium', 'low'])
-          .optional()
-          .describe(
-            'The quality of the image. One of auto (default), high, medium, or low. High/medium/low only supported for gpt-image-1.',
-          ),
-        size: z
-          .enum(['auto', '1024x1024', '1536x1024', '1024x1536', '256x256', '512x512'])
-          .optional()
-          .describe(
-            'The size of the generated images. For gpt-image-1: auto (default), 1024x1024, 1536x1024, 1024x1536. For dall-e-2: 256x256, 512x512, 1024x1024.',
-          ),
-      }),
-      responseFormat: 'content_and_artifact',
-    },
+    oaiToolkit.image_edit_oai,
   );
 
   return [imageGenTool, imageEditTool];

api/app/clients/tools/structured/StableDiffusion.js

@@ -11,14 +11,14 @@ const paths = require('~/config/paths');
 const { logger } = require('~/config');
 
 const displayMessage =
-  'Stable Diffusion displayed an image. All generated images are already plainly visible, so don\'t repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.';
+  "Stable Diffusion displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
 
 class StableDiffusionAPI extends Tool {
   constructor(fields) {
     super();
     /** @type {string} User ID */
     this.userId = fields.userId;
-    /** @type {Express.Request | undefined} Express Request object, only provided by ToolService */
+    /** @type {ServerRequest | undefined} Express Request object, only provided by ToolService */
     this.req = fields.req;
     /** @type {boolean} Used to initialize the Tool without necessary variables. */
     this.override = fields.override ?? false;
@@ -44,7 +44,7 @@ class StableDiffusionAPI extends Tool {
 // "negative_prompt":"semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, out of frame, low quality, ugly, mutation, deformed"
 // - Generate images only once per human query unless explicitly requested by the user`;
     this.description =
-      'You can generate images using text with \'stable-diffusion\'. This tool is exclusively for visual content.';
+      "You can generate images using text with 'stable-diffusion'. This tool is exclusively for visual content.";
     this.schema = z.object({
       prompt: z
         .string()

api/app/clients/tools/structured/YouTube.js

@@ -1,9 +1,9 @@
-const { z } = require('zod');
+const { ytToolkit } = require('@librechat/api');
 const { tool } = require('@langchain/core/tools');
 const { youtube } = require('@googleapis/youtube');
+const { logger } = require('@librechat/data-schemas');
 const { YoutubeTranscript } = require('youtube-transcript');
 const { getApiKey } = require('./credentials');
-const { logger } = require('~/config');
 
 function extractVideoId(url) {
   const rawIdRegex = /^[a-zA-Z0-9_-]{11}$/;
@@ -29,7 +29,7 @@ function parseTranscript(transcriptResponse) {
     .map((entry) => entry.text.trim())
     .filter((text) => text)
     .join(' ')
-    .replaceAll(''', '\'');
+    .replaceAll(''', "'");
 }
 
 function createYouTubeTools(fields = {}) {
@@ -42,160 +42,94 @@ function createYouTubeTools(fields = {}) {
     auth: apiKey,
   });
 
-  const searchTool = tool(
-    async ({ query, maxResults = 5 }) => {
-      const response = await youtubeClient.search.list({
-        part: 'snippet',
-        q: query,
-        type: 'video',
-        maxResults: maxResults || 5,
-      });
-      const result = response.data.items.map((item) => ({
-        title: item.snippet.title,
-        description: item.snippet.description,
-        url: `https://www.youtube.com/watch?v=${item.id.videoId}`,
-      }));
-      return JSON.stringify(result, null, 2);
-    },
-    {
-      name: 'youtube_search',
-      description: `Search for YouTube videos by keyword or phrase.
-- Required: query (search terms to find videos)
-- Optional: maxResults (number of videos to return, 1-50, default: 5)
-- Returns: List of videos with titles, descriptions, and URLs
-- Use for: Finding specific videos, exploring content, research
-Example: query="cooking pasta tutorials" maxResults=3`,
-      schema: z.object({
-        query: z.string().describe('Search query terms'),
-        maxResults: z.number().int().min(1).max(50).optional().describe('Number of results (1-50)'),
-      }),
-    },
-  );
+  const searchTool = tool(async ({ query, maxResults = 5 }) => {
+    const response = await youtubeClient.search.list({
+      part: 'snippet',
+      q: query,
+      type: 'video',
+      maxResults: maxResults || 5,
+    });
+    const result = response.data.items.map((item) => ({
+      title: item.snippet.title,
+      description: item.snippet.description,
+      url: `https://www.youtube.com/watch?v=${item.id.videoId}`,
+    }));
+    return JSON.stringify(result, null, 2);
+  }, ytToolkit.youtube_search);
 
-  const infoTool = tool(
-    async ({ url }) => {
-      const videoId = extractVideoId(url);
-      if (!videoId) {
-        throw new Error('Invalid YouTube URL or video ID');
-      }
+  const infoTool = tool(async ({ url }) => {
+    const videoId = extractVideoId(url);
+    if (!videoId) {
+      throw new Error('Invalid YouTube URL or video ID');
+    }
 
-      const response = await youtubeClient.videos.list({
-        part: 'snippet,statistics',
-        id: videoId,
-      });
+    const response = await youtubeClient.videos.list({
+      part: 'snippet,statistics',
+      id: videoId,
+    });
 
-      if (!response.data.items?.length) {
-        throw new Error('Video not found');
-      }
-      const video = response.data.items[0];
+    if (!response.data.items?.length) {
+      throw new Error('Video not found');
+    }
+    const video = response.data.items[0];
 
-      const result = {
-        title: video.snippet.title,
-        description: video.snippet.description,
-        views: video.statistics.viewCount,
-        likes: video.statistics.likeCount,
-        comments: video.statistics.commentCount,
-      };
-      return JSON.stringify(result, null, 2);
-    },
-    {
-      name: 'youtube_info',
-      description: `Get detailed metadata and statistics for a specific YouTube video.
-- Required: url (full YouTube URL or video ID)
-- Returns: Video title, description, view count, like count, comment count
-- Use for: Getting video metrics and basic metadata
-- DO NOT USE FOR VIDEO SUMMARIES, USE TRANSCRIPTS FOR COMPREHENSIVE ANALYSIS
-- Accepts both full URLs and video IDs
-Example: url="https://youtube.com/watch?v=abc123" or url="abc123"`,
-      schema: z.object({
-        url: z.string().describe('YouTube video URL or ID'),
-      }),
-    },
-  );
+    const result = {
+      title: video.snippet.title,
+      description: video.snippet.description,
+      views: video.statistics.viewCount,
+      likes: video.statistics.likeCount,
+      comments: video.statistics.commentCount,
+    };
+    return JSON.stringify(result, null, 2);
+  }, ytToolkit.youtube_info);
 
-  const commentsTool = tool(
-    async ({ url, maxResults = 10 }) => {
-      const videoId = extractVideoId(url);
-      if (!videoId) {
-        throw new Error('Invalid YouTube URL or video ID');
-      }
+  const commentsTool = tool(async ({ url, maxResults = 10 }) => {
+    const videoId = extractVideoId(url);
+    if (!videoId) {
+      throw new Error('Invalid YouTube URL or video ID');
+    }
 
-      const response = await youtubeClient.commentThreads.list({
-        part: 'snippet',
-        videoId,
-        maxResults: maxResults || 10,
-      });
+    const response = await youtubeClient.commentThreads.list({
+      part: 'snippet',
+      videoId,
+      maxResults: maxResults || 10,
+    });
 
-      const result = response.data.items.map((item) => ({
-        author: item.snippet.topLevelComment.snippet.authorDisplayName,
-        text: item.snippet.topLevelComment.snippet.textDisplay,
-        likes: item.snippet.topLevelComment.snippet.likeCount,
-      }));
-      return JSON.stringify(result, null, 2);
-    },
-    {
-      name: 'youtube_comments',
-      description: `Retrieve top-level comments from a YouTube video.
-- Required: url (full YouTube URL or video ID)
-- Optional: maxResults (number of comments, 1-50, default: 10)
-- Returns: Comment text, author names, like counts
-- Use for: Sentiment analysis, audience feedback, engagement review
-Example: url="abc123" maxResults=20`,
-      schema: z.object({
-        url: z.string().describe('YouTube video URL or ID'),
-        maxResults: z
-          .number()
-          .int()
-          .min(1)
-          .max(50)
-          .optional()
-          .describe('Number of comments to retrieve'),
-      }),
-    },
-  );
+    const result = response.data.items.map((item) => ({
+      author: item.snippet.topLevelComment.snippet.authorDisplayName,
+      text: item.snippet.topLevelComment.snippet.textDisplay,
+      likes: item.snippet.topLevelComment.snippet.likeCount,
+    }));
+    return JSON.stringify(result, null, 2);
+  }, ytToolkit.youtube_comments);
 
-  const transcriptTool = tool(
-    async ({ url }) => {
-      const videoId = extractVideoId(url);
-      if (!videoId) {
-        throw new Error('Invalid YouTube URL or video ID');
+  const transcriptTool = tool(async ({ url }) => {
+    const videoId = extractVideoId(url);
+    if (!videoId) {
+      throw new Error('Invalid YouTube URL or video ID');
+    }
+
+    try {
+      try {
+        const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'en' });
+        return parseTranscript(transcript);
+      } catch (e) {
+        logger.error(e);
       }
 
       try {
-        try {
-          const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'en' });
-          return parseTranscript(transcript);
-        } catch (e) {
-          logger.error(e);
-        }
-
-        try {
-          const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'de' });
-          return parseTranscript(transcript);
-        } catch (e) {
-          logger.error(e);
-        }
-
-        const transcript = await YoutubeTranscript.fetchTranscript(videoId);
+        const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'de' });
         return parseTranscript(transcript);
-      } catch (error) {
-        throw new Error(`Failed to fetch transcript: ${error.message}`);
+      } catch (e) {
+        logger.error(e);
       }
-    },
-    {
-      name: 'youtube_transcript',
-      description: `Fetch and parse the transcript/captions of a YouTube video.
-- Required: url (full YouTube URL or video ID)
-- Returns: Full video transcript as plain text
-- Use for: Content analysis, summarization, translation reference
-- This is the "Go-to" tool for analyzing actual video content
-- Attempts to fetch English first, then German, then any available language
-Example: url="https://youtube.com/watch?v=abc123"`,
-      schema: z.object({
-        url: z.string().describe('YouTube video URL or ID'),
-      }),
-    },
-  );
+
+      const transcript = await YoutubeTranscript.fetchTranscript(videoId);
+      return parseTranscript(transcript);
+    } catch (error) {
+      throw new Error(`Failed to fetch transcript: ${error.message}`);
+    }
+  }, ytToolkit.youtube_transcript);
 
   return [searchTool, infoTool, commentsTool, transcriptTool];
 }

api/app/clients/tools/structured/specs/DALLE3-proxy.spec.js

@@ -0,0 +1,60 @@
+const DALLE3 = require('../DALLE3');
+const { ProxyAgent } = require('undici');
+
+jest.mock('tiktoken');
+const processFileURL = jest.fn();
+
+describe('DALLE3 Proxy Configuration', () => {
+  let originalEnv;
+
+  beforeAll(() => {
+    originalEnv = { ...process.env };
+  });
+
+  beforeEach(() => {
+    jest.resetModules();
+    process.env = { ...originalEnv };
+  });
+
+  afterEach(() => {
+    process.env = originalEnv;
+  });
+
+  it('should configure ProxyAgent in fetchOptions.dispatcher when PROXY env is set', () => {
+    // Set proxy environment variable
+    process.env.PROXY = 'http://proxy.example.com:8080';
+    process.env.DALLE_API_KEY = 'test-api-key';
+
+    // Create instance
+    const dalleWithProxy = new DALLE3({ processFileURL });
+
+    // Check that the openai client exists
+    expect(dalleWithProxy.openai).toBeDefined();
+
+    // Check that _options exists and has fetchOptions with a dispatcher
+    expect(dalleWithProxy.openai._options).toBeDefined();
+    expect(dalleWithProxy.openai._options.fetchOptions).toBeDefined();
+    expect(dalleWithProxy.openai._options.fetchOptions.dispatcher).toBeDefined();
+    expect(dalleWithProxy.openai._options.fetchOptions.dispatcher).toBeInstanceOf(ProxyAgent);
+  });
+
+  it('should not configure ProxyAgent when PROXY env is not set', () => {
+    // Ensure PROXY is not set
+    delete process.env.PROXY;
+    process.env.DALLE_API_KEY = 'test-api-key';
+
+    // Create instance
+    const dalleWithoutProxy = new DALLE3({ processFileURL });
+
+    // Check that the openai client exists
+    expect(dalleWithoutProxy.openai).toBeDefined();
+
+    // Check that _options exists but fetchOptions either doesn't exist or doesn't have a dispatcher
+    expect(dalleWithoutProxy.openai._options).toBeDefined();
+
+    // fetchOptions should either not exist or not have a dispatcher
+    if (dalleWithoutProxy.openai._options.fetchOptions) {
+      expect(dalleWithoutProxy.openai._options.fetchOptions.dispatcher).toBeUndefined();
+    }
+  });
+});

api/app/clients/tools/structured/specs/DALLE3.spec.js

@@ -1,9 +1,8 @@
 const OpenAI = require('openai');
+const { logger } = require('@librechat/data-schemas');
 const DALLE3 = require('../DALLE3');
-const logger = require('~/config/winston');
 
 jest.mock('openai');
-
 jest.mock('@librechat/data-schemas', () => {
   return {
     logger: {
@@ -26,25 +25,6 @@ jest.mock('tiktoken', () => {
 
 const processFileURL = jest.fn();
 
-jest.mock('~/server/services/Files/images', () => ({
-  getImageBasename: jest.fn().mockImplementation((url) => {
-    // Split the URL by '/'
-    const parts = url.split('/');
-
-    // Get the last part of the URL
-    const lastPart = parts.pop();
-
-    // Check if the last part of the URL matches the image extension regex
-    const imageExtensionRegex = /\.(jpg|jpeg|png|gif|bmp|tiff|svg)$/i;
-    if (imageExtensionRegex.test(lastPart)) {
-      return lastPart;
-    }
-
-    // If the regex test fails, return an empty string
-    return '';
-  }),
-}));
-
 const generate = jest.fn();
 OpenAI.mockImplementation(() => ({
   images: {

api/app/clients/tools/util/fileSearch.js

@@ -1,26 +1,46 @@
 const { z } = require('zod');
 const axios = require('axios');
 const { tool } = require('@langchain/core/tools');
+const { logger } = require('@librechat/data-schemas');
+const { generateShortLivedToken } = require('@librechat/api');
 const { Tools, EToolResources } = require('librechat-data-provider');
+const { filterFilesByAgentAccess } = require('~/server/services/Files/permissions');
 const { getFiles } = require('~/models/File');
-const { logger } = require('~/config');
 
 /**
  *
  * @param {Object} options
  * @param {ServerRequest} options.req
  * @param {Agent['tool_resources']} options.tool_resources
+ * @param {string} [options.agentId] - The agent ID for file access control
  * @returns {Promise<{
  *   files: Array<{ file_id: string; filename: string }>,
  *   toolContext: string
  * }>}
  */
 const primeFiles = async (options) => {
-  const { tool_resources } = options;
+  const { tool_resources, req, agentId } = options;
   const file_ids = tool_resources?.[EToolResources.file_search]?.file_ids ?? [];
   const agentResourceIds = new Set(file_ids);
   const resourceFiles = tool_resources?.[EToolResources.file_search]?.files ?? [];
-  const dbFiles = ((await getFiles({ file_id: { $in: file_ids } })) ?? []).concat(resourceFiles);
+
+  // Get all files first
+  const allFiles = (await getFiles({ file_id: { $in: file_ids } }, null, { text: 0 })) ?? [];
+
+  // Filter by access if user and agent are provided
+  let dbFiles;
+  if (req?.user?.id && agentId) {
+    dbFiles = await filterFilesByAgentAccess({
+      files: allFiles,
+      userId: req.user.id,
+      role: req.user.role,
+      agentId,
+    });
+  } else {
+    dbFiles = allFiles;
+  }
+
+  dbFiles = dbFiles.concat(resourceFiles);
 
   let toolContext = `- Note: Semantic search is available through the ${Tools.file_search} tool but no files are currently loaded. Request the user to upload documents to search through.`;
 
@@ -51,15 +71,16 @@ const primeFiles = async (options) => {
  * @param {ServerRequest} options.req
  * @param {Array<{ file_id: string; filename: string }>} options.files
  * @param {string} [options.entity_id]
+ * @param {boolean} [options.fileCitations=false] - Whether to include citation instructions
  * @returns
  */
-const createFileSearchTool = async ({ req, files, entity_id }) => {
+const createFileSearchTool = async ({ req, files, entity_id, fileCitations = false }) => {
   return tool(
     async ({ query }) => {
       if (files.length === 0) {
         return 'No files to search. Instruct the user to add files for the search.';
       }
-      const jwtToken = req.headers.authorization.split(' ')[1];
+      const jwtToken = generateShortLivedToken(req.user.id);
       if (!jwtToken) {
         return 'There was an error authenticating the file search request.';
       }
@@ -105,11 +126,13 @@ const createFileSearchTool = async ({ req, files, entity_id }) => {
       }
 
       const formattedResults = validResults
-        .flatMap((result) =>
+        .flatMap((result, fileIndex) =>
           result.data.map(([docInfo, distance]) => ({
             filename: docInfo.metadata.source.split('/').pop(),
             content: docInfo.page_content,
             distance,
+            file_id: files[fileIndex]?.file_id,
+            page: docInfo.metadata.page || null,
           })),
         )
         // TODO: results should be sorted by relevance, not distance
@@ -119,18 +142,41 @@ const createFileSearchTool = async ({ req, files, entity_id }) => {
 
       const formattedString = formattedResults
         .map(
-          (result) =>
-            `File: ${result.filename}\nRelevance: ${1.0 - result.distance.toFixed(4)}\nContent: ${
-              result.content
-            }\n`,
+          (result, index) =>
+            `File: ${result.filename}${
+              fileCitations ? `\nAnchor: \\ue202turn0file${index} (${result.filename})` : ''
+            }\nRelevance: ${(1.0 - result.distance).toFixed(4)}\nContent: ${result.content}\n`,
         )
         .join('\n---\n');
 
-      return formattedString;
+      const sources = formattedResults.map((result) => ({
+        type: 'file',
+        fileId: result.file_id,
+        content: result.content,
+        fileName: result.filename,
+        relevance: 1.0 - result.distance,
+        pages: result.page ? [result.page] : [],
+        pageRelevance: result.page ? { [result.page]: 1.0 - result.distance } : {},
+      }));
+
+      return [formattedString, { [Tools.file_search]: { sources, fileCitations } }];
     },
     {
       name: Tools.file_search,
-      description: `Performs semantic search across attached "${Tools.file_search}" documents using natural language queries. This tool analyzes the content of uploaded files to find relevant information, quotes, and passages that best match your query. Use this to extract specific information or find relevant sections within the available documents.`,
+      responseFormat: 'content_and_artifact',
+      description: `Performs semantic search across attached "${Tools.file_search}" documents using natural language queries. This tool analyzes the content of uploaded files to find relevant information, quotes, and passages that best match your query. Use this to extract specific information or find relevant sections within the available documents.${
+        fileCitations
+          ? `
+
+**CITE FILE SEARCH RESULTS:**
+Use anchor markers immediately after statements derived from file content. Reference the filename in your text:
+- File citation: "The document.pdf states that... \\ue202turn0file0"  
+- Page reference: "According to report.docx... \\ue202turn0file1"
+- Multi-file: "Multiple sources confirm... \\ue200\\ue202turn0file0\\ue202turn0file1\\ue201"
+
+**ALWAYS mention the filename in your text before the citation marker. NEVER use markdown links or footnotes.**`
+          : ''
+      }`,
       schema: z.object({
         query: z
           .string()

api/app/clients/tools/util/handleTools.js

@@ -1,14 +1,16 @@
+const { logger } = require('@librechat/data-schemas');
 const { SerpAPI } = require('@langchain/community/tools/serpapi');
 const { Calculator } = require('@langchain/community/tools/calculator');
+const { mcpToolPattern, loadWebSearchAuth, checkAccess } = require('@librechat/api');
 const { EnvVar, createCodeExecutionTool, createSearchTool } = require('@librechat/agents');
 const {
   Tools,
   Constants,
+  Permissions,
   EToolResources,
-  loadWebSearchAuth,
+  PermissionTypes,
   replaceSpecialVars,
 } = require('librechat-data-provider');
-const { getUserPluginAuthValue } = require('~/server/services/PluginService');
 const {
   availableTools,
   manifestToolMap,
@@ -28,11 +30,11 @@ const {
 } = require('../');
 const { primeFiles: primeCodeFiles } = require('~/server/services/Files/Code/process');
 const { createFileSearchTool, primeFiles: primeSearchFiles } = require('./fileSearch');
+const { getUserPluginAuthValue } = require('~/server/services/PluginService');
+const { createMCPTool, createMCPTools } = require('~/server/services/MCP');
 const { loadAuthValues } = require('~/server/services/Tools/credentials');
-const { createMCPTool } = require('~/server/services/MCP');
-const { logger } = require('~/config');
-
-const mcpToolPattern = new RegExp(`^.+${Constants.mcp_delimiter}.+$`);
+const { getCachedTools } = require('~/server/services/Config');
+const { getRoleByName } = require('~/models/Role');
 
 /**
  * Validates the availability and authentication of tools for a user based on environment variables or user-specific plugin authentication values.
@@ -93,7 +95,7 @@ const validateTools = async (user, tools = []) => {
     return Array.from(validToolsSet.values());
   } catch (err) {
     logger.error('[validateTools] There was a problem validating tools', err);
-    throw new Error('There was a problem validating tools');
+    throw new Error(err);
   }
 };
 
@@ -127,27 +129,37 @@ const getAuthFields = (toolKey) => {
 
 /**
  *
- * @param {object} object
- * @param {string} object.user
- * @param {Pick<Agent, 'id' | 'provider' | 'model'>} [object.agent]
- * @param {string} [object.model]
- * @param {EModelEndpoint} [object.endpoint]
- * @param {LoadToolOptions} [object.options]
- * @param {boolean} [object.useSpecs]
- * @param {Array<string>} object.tools
- * @param {boolean} [object.functions]
- * @param {boolean} [object.returnMap]
+ * @param {object} params
+ * @param {string} params.user
+ * @param {Record<string, Record<string, string>>} [object.userMCPAuthMap]
+ * @param {AbortSignal} [object.signal]
+ * @param {Pick<Agent, 'id' | 'provider' | 'model'>} [params.agent]
+ * @param {string} [params.model]
+ * @param {EModelEndpoint} [params.endpoint]
+ * @param {LoadToolOptions} [params.options]
+ * @param {boolean} [params.useSpecs]
+ * @param {Array<string>} params.tools
+ * @param {boolean} [params.functions]
+ * @param {boolean} [params.returnMap]
+ * @param {AppConfig['webSearch']} [params.webSearch]
+ * @param {AppConfig['fileStrategy']} [params.fileStrategy]
+ * @param {AppConfig['imageOutputType']} [params.imageOutputType]
  * @returns {Promise<{ loadedTools: Tool[], toolContextMap: Object<string, any> } | Record<string,Tool>>}
  */
 const loadTools = async ({
   user,
   agent,
   model,
+  signal,
   endpoint,
+  userMCPAuthMap,
   tools = [],
   options = {},
   functions = true,
   returnMap = false,
+  webSearch,
+  fileStrategy,
+  imageOutputType,
 }) => {
   const toolConstructors = {
     flux: FluxAPI,
@@ -206,6 +218,8 @@ const loadTools = async ({
         ...authValues,
         isAgent: !!agent,
         req: options.req,
+        imageOutputType,
+        fileStrategy,
         imageFiles,
       });
     },
@@ -221,7 +235,7 @@ const loadTools = async ({
   const imageGenOptions = {
     isAgent: !!agent,
     req: options.req,
-    fileStrategy: options.fileStrategy,
+    fileStrategy,
     processFileURL: options.processFileURL,
     returnMetadata: options.returnMetadata,
     uploadImageBuffer: options.uploadImageBuffer,
@@ -236,7 +250,8 @@ const loadTools = async ({
 
   /** @type {Record<string, string>} */
   const toolContextMap = {};
-  const appTools = options.req?.app?.locals?.availableTools ?? {};
+  const cachedTools = (await getCachedTools({ userId: user, includeGlobal: true })) ?? {};
+  const requestedMCPTools = {};
 
   for (const tool of tools) {
     if (tool === Tools.execute_code) {
@@ -246,7 +261,13 @@ const loadTools = async ({
           authFields: [EnvVar.CODE_API_KEY],
         });
         const codeApiKey = authValues[EnvVar.CODE_API_KEY];
-        const { files, toolContext } = await primeCodeFiles(options, codeApiKey);
+        const { files, toolContext } = await primeCodeFiles(
+          {
+            ...options,
+            agentId: agent?.id,
+          },
+          codeApiKey,
+        );
         if (toolContext) {
           toolContextMap[tool] = toolContext;
         }
@@ -261,19 +282,43 @@ const loadTools = async ({
       continue;
     } else if (tool === Tools.file_search) {
       requestedTools[tool] = async () => {
-        const { files, toolContext } = await primeSearchFiles(options);
+        const { files, toolContext } = await primeSearchFiles({
+          ...options,
+          agentId: agent?.id,
+        });
         if (toolContext) {
           toolContextMap[tool] = toolContext;
         }
-        return createFileSearchTool({ req: options.req, files, entity_id: agent?.id });
+
+        /** @type {boolean | undefined} Check if user has FILE_CITATIONS permission */
+        let fileCitations;
+        if (fileCitations == null && options.req?.user != null) {
+          try {
+            fileCitations = await checkAccess({
+              user: options.req.user,
+              permissionType: PermissionTypes.FILE_CITATIONS,
+              permissions: [Permissions.USE],
+              getRoleByName,
+            });
+          } catch (error) {
+            logger.error('[handleTools] FILE_CITATIONS permission check failed:', error);
+            fileCitations = false;
+          }
+        }
+
+        return createFileSearchTool({
+          req: options.req,
+          files,
+          entity_id: agent?.id,
+          fileCitations,
+        });
       };
       continue;
     } else if (tool === Tools.web_search) {
-      const webSearchConfig = options?.req?.app?.locals?.webSearch;
       const result = await loadWebSearchAuth({
         userId: user,
         loadAuthValues,
-        webSearchConfig,
+        webSearchConfig: webSearch,
       });
       const { onSearchResults, onGetHighlights } = options?.[Tools.web_search] ?? {};
       requestedTools[tool] = async () => {
@@ -295,14 +340,46 @@ Current Date & Time: ${replaceSpecialVars({ text: '{{iso_datetime}}' })}
         });
       };
       continue;
-    } else if (tool && appTools[tool] && mcpToolPattern.test(tool)) {
-      requestedTools[tool] = async () =>
+    } else if (tool && cachedTools && mcpToolPattern.test(tool)) {
+      const [toolName, serverName] = tool.split(Constants.mcp_delimiter);
+      if (toolName === Constants.mcp_server) {
+        /** Placeholder used for UI purposes */
+        continue;
+      }
+      if (serverName && options.req?.config?.mcpConfig?.[serverName] == null) {
+        logger.warn(
+          `MCP server "${serverName}" for "${toolName}" tool is not configured${agent?.id != null && agent.id ? ` but attached to "${agent.id}"` : ''}`,
+        );
+        continue;
+      }
+      if (toolName === Constants.mcp_all) {
+        const currentMCPGenerator = async (index) =>
+          createMCPTools({
+            req: options.req,
+            res: options.res,
+            index,
+            serverName,
+            userMCPAuthMap,
+            model: agent?.model ?? model,
+            provider: agent?.provider ?? endpoint,
+            signal,
+          });
+        requestedMCPTools[serverName] = [currentMCPGenerator];
+        continue;
+      }
+      const currentMCPGenerator = async (index) =>
         createMCPTool({
+          index,
           req: options.req,
+          res: options.res,
           toolKey: tool,
+          userMCPAuthMap,
           model: agent?.model ?? model,
           provider: agent?.provider ?? endpoint,
+          signal,
         });
+      requestedMCPTools[serverName] = requestedMCPTools[serverName] || [];
+      requestedMCPTools[serverName].push(currentMCPGenerator);
       continue;
     }
 
@@ -342,6 +419,34 @@ Current Date & Time: ${replaceSpecialVars({ text: '{{iso_datetime}}' })}
   }
 
   const loadedTools = (await Promise.all(toolPromises)).flatMap((plugin) => plugin || []);
+  const mcpToolPromises = [];
+  /** MCP server tools are initialized sequentially by server */
+  let index = -1;
+  for (const [serverName, generators] of Object.entries(requestedMCPTools)) {
+    index++;
+    for (const generator of generators) {
+      try {
+        if (generator && generators.length === 1) {
+          mcpToolPromises.push(
+            generator(index).catch((error) => {
+              logger.error(`Error loading ${serverName} tools:`, error);
+              return null;
+            }),
+          );
+          continue;
+        }
+        const mcpTool = await generator(index);
+        if (Array.isArray(mcpTool)) {
+          loadedTools.push(...mcpTool);
+        } else if (mcpTool) {
+          loadedTools.push(mcpTool);
+        }
+      } catch (error) {
+        logger.error(`Error loading MCP tool for server ${serverName}:`, error);
+      }
+    }
+  }
+  loadedTools.push(...(await Promise.all(mcpToolPromises)).flatMap((plugin) => plugin || []));
   return { loadedTools, toolContextMap };
 };
 

api/app/clients/tools/util/handleTools.test.js

@@ -9,6 +9,27 @@ const mockPluginService = {
 
 jest.mock('~/server/services/PluginService', () => mockPluginService);
 
+jest.mock('~/server/services/Config', () => ({
+  getAppConfig: jest.fn().mockResolvedValue({
+    // Default app config for tool tests
+    paths: { uploads: '/tmp' },
+    fileStrategy: 'local',
+    filteredTools: [],
+    includedTools: [],
+  }),
+  getCachedTools: jest.fn().mockResolvedValue({
+    // Default cached tools for tests
+    dalle: {
+      type: 'function',
+      function: {
+        name: 'dalle',
+        description: 'DALL-E image generation',
+        parameters: {},
+      },
+    },
+  }),
+}));
+
 const { BaseLLM } = require('@langchain/openai');
 const { Calculator } = require('@langchain/community/tools/calculator');
 

api/cache/banViolation.js

@@ -1,7 +1,8 @@
 const { logger } = require('@librechat/data-schemas');
+const { isEnabled, math } = require('@librechat/api');
 const { ViolationTypes } = require('librechat-data-provider');
-const { isEnabled, math, removePorts } = require('~/server/utils');
 const { deleteAllUserSessions } = require('~/models');
+const { removePorts } = require('~/server/utils');
 const getLogStores = require('./getLogStores');
 
 const { BAN_VIOLATIONS, BAN_INTERVAL } = process.env ?? {};

api/cache/cacheConfig.js

@@ -0,0 +1,66 @@
+const fs = require('fs');
+const { math, isEnabled } = require('@librechat/api');
+const { CacheKeys } = require('librechat-data-provider');
+
+// To ensure that different deployments do not interfere with each other's cache, we use a prefix for the Redis keys.
+// This prefix is usually the deployment ID, which is often passed to the container or pod as an env var.
+// Set REDIS_KEY_PREFIX_VAR to the env var that contains the deployment ID.
+const REDIS_KEY_PREFIX_VAR = process.env.REDIS_KEY_PREFIX_VAR;
+const REDIS_KEY_PREFIX = process.env.REDIS_KEY_PREFIX;
+if (REDIS_KEY_PREFIX_VAR && REDIS_KEY_PREFIX) {
+  throw new Error('Only either REDIS_KEY_PREFIX_VAR or REDIS_KEY_PREFIX can be set.');
+}
+
+const USE_REDIS = isEnabled(process.env.USE_REDIS);
+if (USE_REDIS && !process.env.REDIS_URI) {
+  throw new Error('USE_REDIS is enabled but REDIS_URI is not set.');
+}
+
+// Comma-separated list of cache namespaces that should be forced to use in-memory storage
+// even when Redis is enabled. This allows selective performance optimization for specific caches.
+const FORCED_IN_MEMORY_CACHE_NAMESPACES = process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES
+  ? process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES.split(',').map((key) => key.trim())
+  : [];
+
+// Validate against CacheKeys enum
+if (FORCED_IN_MEMORY_CACHE_NAMESPACES.length > 0) {
+  const validKeys = Object.values(CacheKeys);
+  const invalidKeys = FORCED_IN_MEMORY_CACHE_NAMESPACES.filter((key) => !validKeys.includes(key));
+
+  if (invalidKeys.length > 0) {
+    throw new Error(
+      `Invalid cache keys in FORCED_IN_MEMORY_CACHE_NAMESPACES: ${invalidKeys.join(', ')}. Valid keys: ${validKeys.join(', ')}`,
+    );
+  }
+}
+
+const cacheConfig = {
+  FORCED_IN_MEMORY_CACHE_NAMESPACES,
+  USE_REDIS,
+  REDIS_URI: process.env.REDIS_URI,
+  REDIS_USERNAME: process.env.REDIS_USERNAME,
+  REDIS_PASSWORD: process.env.REDIS_PASSWORD,
+  REDIS_CA: process.env.REDIS_CA ? fs.readFileSync(process.env.REDIS_CA, 'utf8') : null,
+  REDIS_KEY_PREFIX: process.env[REDIS_KEY_PREFIX_VAR] || REDIS_KEY_PREFIX || '',
+  REDIS_MAX_LISTENERS: math(process.env.REDIS_MAX_LISTENERS, 40),
+  REDIS_PING_INTERVAL: math(process.env.REDIS_PING_INTERVAL, 0),
+  /** Max delay between reconnection attempts in ms */
+  REDIS_RETRY_MAX_DELAY: math(process.env.REDIS_RETRY_MAX_DELAY, 3000),
+  /** Max number of reconnection attempts (0 = infinite) */
+  REDIS_RETRY_MAX_ATTEMPTS: math(process.env.REDIS_RETRY_MAX_ATTEMPTS, 10),
+  /** Connection timeout in ms */
+  REDIS_CONNECT_TIMEOUT: math(process.env.REDIS_CONNECT_TIMEOUT, 10000),
+  /** Queue commands when disconnected */
+  REDIS_ENABLE_OFFLINE_QUEUE: isEnabled(process.env.REDIS_ENABLE_OFFLINE_QUEUE ?? 'true'),
+  /** flag to modify redis connection by adding dnsLookup this is required when connecting to elasticache for ioredis
+   * see "Special Note: Aws Elasticache Clusters with TLS" on this webpage:  https://www.npmjs.com/package/ioredis **/
+  REDIS_USE_ALTERNATIVE_DNS_LOOKUP: isEnabled(process.env.REDIS_USE_ALTERNATIVE_DNS_LOOKUP),
+  /** Enable redis cluster without the need of multiple URIs */
+  USE_REDIS_CLUSTER: isEnabled(process.env.USE_REDIS_CLUSTER ?? 'false'),
+  CI: isEnabled(process.env.CI),
+  DEBUG_MEMORY_CACHE: isEnabled(process.env.DEBUG_MEMORY_CACHE),
+
+  BAN_DURATION: math(process.env.BAN_DURATION, 7200000), // 2 hours
+};
+
+module.exports = { cacheConfig };

api/cache/cacheConfig.spec.js

@@ -0,0 +1,189 @@
+const fs = require('fs');
+
+describe('cacheConfig', () => {
+  let originalEnv;
+  let originalReadFileSync;
+
+  beforeEach(() => {
+    originalEnv = { ...process.env };
+    originalReadFileSync = fs.readFileSync;
+
+    // Clear all related env vars first
+    delete process.env.REDIS_URI;
+    delete process.env.REDIS_CA;
+    delete process.env.REDIS_KEY_PREFIX_VAR;
+    delete process.env.REDIS_KEY_PREFIX;
+    delete process.env.USE_REDIS;
+    delete process.env.USE_REDIS_CLUSTER;
+    delete process.env.REDIS_PING_INTERVAL;
+    delete process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES;
+
+    // Clear require cache
+    jest.resetModules();
+  });
+
+  afterEach(() => {
+    process.env = originalEnv;
+    fs.readFileSync = originalReadFileSync;
+    jest.resetModules();
+  });
+
+  describe('REDIS_KEY_PREFIX validation and resolution', () => {
+    test('should throw error when both REDIS_KEY_PREFIX_VAR and REDIS_KEY_PREFIX are set', () => {
+      process.env.REDIS_KEY_PREFIX_VAR = 'DEPLOYMENT_ID';
+      process.env.REDIS_KEY_PREFIX = 'manual-prefix';
+
+      expect(() => {
+        require('./cacheConfig');
+      }).toThrow('Only either REDIS_KEY_PREFIX_VAR or REDIS_KEY_PREFIX can be set.');
+    });
+
+    test('should resolve REDIS_KEY_PREFIX from variable reference', () => {
+      process.env.REDIS_KEY_PREFIX_VAR = 'DEPLOYMENT_ID';
+      process.env.DEPLOYMENT_ID = 'test-deployment-123';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_KEY_PREFIX).toBe('test-deployment-123');
+    });
+
+    test('should use direct REDIS_KEY_PREFIX value', () => {
+      process.env.REDIS_KEY_PREFIX = 'direct-prefix';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_KEY_PREFIX).toBe('direct-prefix');
+    });
+
+    test('should default to empty string when no prefix is configured', () => {
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_KEY_PREFIX).toBe('');
+    });
+
+    test('should handle empty variable reference', () => {
+      process.env.REDIS_KEY_PREFIX_VAR = 'EMPTY_VAR';
+      process.env.EMPTY_VAR = '';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_KEY_PREFIX).toBe('');
+    });
+
+    test('should handle undefined variable reference', () => {
+      process.env.REDIS_KEY_PREFIX_VAR = 'UNDEFINED_VAR';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_KEY_PREFIX).toBe('');
+    });
+  });
+
+  describe('USE_REDIS and REDIS_URI validation', () => {
+    test('should throw error when USE_REDIS is enabled but REDIS_URI is not set', () => {
+      process.env.USE_REDIS = 'true';
+
+      expect(() => {
+        require('./cacheConfig');
+      }).toThrow('USE_REDIS is enabled but REDIS_URI is not set.');
+    });
+
+    test('should not throw error when USE_REDIS is enabled and REDIS_URI is set', () => {
+      process.env.USE_REDIS = 'true';
+      process.env.REDIS_URI = 'redis://localhost:6379';
+
+      expect(() => {
+        require('./cacheConfig');
+      }).not.toThrow();
+    });
+
+    test('should handle empty REDIS_URI when USE_REDIS is enabled', () => {
+      process.env.USE_REDIS = 'true';
+      process.env.REDIS_URI = '';
+
+      expect(() => {
+        require('./cacheConfig');
+      }).toThrow('USE_REDIS is enabled but REDIS_URI is not set.');
+    });
+  });
+
+  describe('USE_REDIS_CLUSTER configuration', () => {
+    test('should default to false when USE_REDIS_CLUSTER is not set', () => {
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.USE_REDIS_CLUSTER).toBe(false);
+    });
+
+    test('should be false when USE_REDIS_CLUSTER is set to false', () => {
+      process.env.USE_REDIS_CLUSTER = 'false';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.USE_REDIS_CLUSTER).toBe(false);
+    });
+
+    test('should be true when USE_REDIS_CLUSTER is set to true', () => {
+      process.env.USE_REDIS_CLUSTER = 'true';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.USE_REDIS_CLUSTER).toBe(true);
+    });
+
+    test('should work with USE_REDIS enabled and REDIS_URI set', () => {
+      process.env.USE_REDIS_CLUSTER = 'true';
+      process.env.USE_REDIS = 'true';
+      process.env.REDIS_URI = 'redis://localhost:6379';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.USE_REDIS_CLUSTER).toBe(true);
+      expect(cacheConfig.USE_REDIS).toBe(true);
+      expect(cacheConfig.REDIS_URI).toBe('redis://localhost:6379');
+    });
+  });
+
+  describe('REDIS_CA file reading', () => {
+    test('should be null when REDIS_CA is not set', () => {
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_CA).toBeNull();
+    });
+  });
+
+  describe('REDIS_PING_INTERVAL configuration', () => {
+    test('should default to 0 when REDIS_PING_INTERVAL is not set', () => {
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_PING_INTERVAL).toBe(0);
+    });
+
+    test('should use provided REDIS_PING_INTERVAL value', () => {
+      process.env.REDIS_PING_INTERVAL = '300';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.REDIS_PING_INTERVAL).toBe(300);
+    });
+  });
+
+  describe('FORCED_IN_MEMORY_CACHE_NAMESPACES validation', () => {
+    test('should parse comma-separated cache keys correctly', () => {
+      process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES = ' ROLES, MESSAGES ';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES).toEqual([
+        'ROLES',
+        'MESSAGES',
+      ]);
+    });
+
+    test('should throw error for invalid cache keys', () => {
+      process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES = 'INVALID_KEY,ROLES';
+
+      expect(() => {
+        require('./cacheConfig');
+      }).toThrow('Invalid cache keys in FORCED_IN_MEMORY_CACHE_NAMESPACES: INVALID_KEY');
+    });
+
+    test('should handle empty string gracefully', () => {
+      process.env.FORCED_IN_MEMORY_CACHE_NAMESPACES = '';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES).toEqual([]);
+    });
+
+    test('should handle undefined env var gracefully', () => {
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES).toEqual([]);
+    });
+  });
+});

api/cache/cacheFactory.js

@@ -0,0 +1,108 @@
+const KeyvRedis = require('@keyv/redis').default;
+const { Keyv } = require('keyv');
+const { RedisStore } = require('rate-limit-redis');
+const { Time } = require('librechat-data-provider');
+const { logger } = require('@librechat/data-schemas');
+const { RedisStore: ConnectRedis } = require('connect-redis');
+const MemoryStore = require('memorystore')(require('express-session'));
+const { keyvRedisClient, ioredisClient, GLOBAL_PREFIX_SEPARATOR } = require('./redisClients');
+const { cacheConfig } = require('./cacheConfig');
+const { violationFile } = require('./keyvFiles');
+
+/**
+ * Creates a cache instance using Redis or a fallback store. Suitable for general caching needs.
+ * @param {string} namespace - The cache namespace.
+ * @param {number} [ttl] - Time to live for cache entries.
+ * @param {object} [fallbackStore] - Optional fallback store if Redis is not used.
+ * @returns {Keyv} Cache instance.
+ */
+const standardCache = (namespace, ttl = undefined, fallbackStore = undefined) => {
+  if (
+    cacheConfig.USE_REDIS &&
+    !cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES?.includes(namespace)
+  ) {
+    try {
+      const keyvRedis = new KeyvRedis(keyvRedisClient);
+      const cache = new Keyv(keyvRedis, { namespace, ttl });
+      keyvRedis.namespace = cacheConfig.REDIS_KEY_PREFIX;
+      keyvRedis.keyPrefixSeparator = GLOBAL_PREFIX_SEPARATOR;
+
+      cache.on('error', (err) => {
+        logger.error(`Cache error in namespace ${namespace}:`, err);
+      });
+
+      return cache;
+    } catch (err) {
+      logger.error(`Failed to create Redis cache for namespace ${namespace}:`, err);
+      throw err;
+    }
+  }
+  if (fallbackStore) return new Keyv({ store: fallbackStore, namespace, ttl });
+  return new Keyv({ namespace, ttl });
+};
+
+/**
+ * Creates a cache instance for storing violation data.
+ * Uses a file-based fallback store if Redis is not enabled.
+ * @param {string} namespace - The cache namespace for violations.
+ * @param {number} [ttl] - Time to live for cache entries.
+ * @returns {Keyv} Cache instance for violations.
+ */
+const violationCache = (namespace, ttl = undefined) => {
+  return standardCache(`violations:${namespace}`, ttl, violationFile);
+};
+
+/**
+ * Creates a session cache instance using Redis or in-memory store.
+ * @param {string} namespace - The session namespace.
+ * @param {number} [ttl] - Time to live for session entries.
+ * @returns {MemoryStore | ConnectRedis} Session store instance.
+ */
+const sessionCache = (namespace, ttl = undefined) => {
+  namespace = namespace.endsWith(':') ? namespace : `${namespace}:`;
+  if (!cacheConfig.USE_REDIS) return new MemoryStore({ ttl, checkPeriod: Time.ONE_DAY });
+  const store = new ConnectRedis({ client: ioredisClient, ttl, prefix: namespace });
+  if (ioredisClient) {
+    ioredisClient.on('error', (err) => {
+      logger.error(`Session store Redis error for namespace ${namespace}:`, err);
+    });
+  }
+  return store;
+};
+
+/**
+ * Creates a rate limiter cache using Redis.
+ * @param {string} prefix - The key prefix for rate limiting.
+ * @returns {RedisStore|undefined} RedisStore instance or undefined if Redis is not used.
+ */
+const limiterCache = (prefix) => {
+  if (!prefix) throw new Error('prefix is required');
+  if (!cacheConfig.USE_REDIS) return undefined;
+  prefix = prefix.endsWith(':') ? prefix : `${prefix}:`;
+
+  try {
+    if (!ioredisClient) {
+      logger.warn(`Redis client not available for rate limiter with prefix ${prefix}`);
+      return undefined;
+    }
+
+    return new RedisStore({ sendCommand, prefix });
+  } catch (err) {
+    logger.error(`Failed to create Redis rate limiter for prefix ${prefix}:`, err);
+    return undefined;
+  }
+};
+
+const sendCommand = (...args) => {
+  if (!ioredisClient) {
+    logger.warn('Redis client not available for command execution');
+    return Promise.reject(new Error('Redis client not available'));
+  }
+
+  return ioredisClient.call(...args).catch((err) => {
+    logger.error('Redis command execution failed:', err);
+    throw err;
+  });
+};
+
+module.exports = { standardCache, sessionCache, violationCache, limiterCache };

api/cache/cacheFactory.spec.js

@@ -0,0 +1,432 @@
+const { Time } = require('librechat-data-provider');
+
+// Mock dependencies first
+const mockKeyvRedis = {
+  namespace: '',
+  keyPrefixSeparator: '',
+};
+
+const mockKeyv = jest.fn().mockReturnValue({
+  mock: 'keyv',
+  on: jest.fn(),
+});
+const mockConnectRedis = jest.fn().mockReturnValue({ mock: 'connectRedis' });
+const mockMemoryStore = jest.fn().mockReturnValue({ mock: 'memoryStore' });
+const mockRedisStore = jest.fn().mockReturnValue({ mock: 'redisStore' });
+
+const mockIoredisClient = {
+  call: jest.fn(),
+  on: jest.fn(),
+};
+
+const mockKeyvRedisClient = {};
+const mockViolationFile = {};
+
+// Mock modules before requiring the main module
+jest.mock('@keyv/redis', () => ({
+  default: jest.fn().mockImplementation(() => mockKeyvRedis),
+}));
+
+jest.mock('keyv', () => ({
+  Keyv: mockKeyv,
+}));
+
+jest.mock('./cacheConfig', () => ({
+  cacheConfig: {
+    USE_REDIS: false,
+    REDIS_KEY_PREFIX: 'test',
+    FORCED_IN_MEMORY_CACHE_NAMESPACES: [],
+  },
+}));
+
+jest.mock('./redisClients', () => ({
+  keyvRedisClient: mockKeyvRedisClient,
+  ioredisClient: mockIoredisClient,
+  GLOBAL_PREFIX_SEPARATOR: '::',
+}));
+
+jest.mock('./keyvFiles', () => ({
+  violationFile: mockViolationFile,
+}));
+
+jest.mock('connect-redis', () => ({ RedisStore: mockConnectRedis }));
+
+jest.mock('memorystore', () => jest.fn(() => mockMemoryStore));
+
+jest.mock('rate-limit-redis', () => ({
+  RedisStore: mockRedisStore,
+}));
+
+jest.mock('@librechat/data-schemas', () => ({
+  logger: {
+    error: jest.fn(),
+    warn: jest.fn(),
+    info: jest.fn(),
+  },
+}));
+
+// Import after mocking
+const { standardCache, sessionCache, violationCache, limiterCache } = require('./cacheFactory');
+const { cacheConfig } = require('./cacheConfig');
+
+describe('cacheFactory', () => {
+  beforeEach(() => {
+    jest.clearAllMocks();
+
+    // Reset cache config mock
+    cacheConfig.USE_REDIS = false;
+    cacheConfig.REDIS_KEY_PREFIX = 'test';
+    cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES = [];
+  });
+
+  describe('redisCache', () => {
+    it('should create Redis cache when USE_REDIS is true', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'test-namespace';
+      const ttl = 3600;
+
+      standardCache(namespace, ttl);
+
+      expect(require('@keyv/redis').default).toHaveBeenCalledWith(mockKeyvRedisClient);
+      expect(mockKeyv).toHaveBeenCalledWith(mockKeyvRedis, { namespace, ttl });
+      expect(mockKeyvRedis.namespace).toBe(cacheConfig.REDIS_KEY_PREFIX);
+      expect(mockKeyvRedis.keyPrefixSeparator).toBe('::');
+    });
+
+    it('should create Redis cache with undefined ttl when not provided', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'test-namespace';
+
+      standardCache(namespace);
+
+      expect(mockKeyv).toHaveBeenCalledWith(mockKeyvRedis, { namespace, ttl: undefined });
+    });
+
+    it('should use fallback store when USE_REDIS is false and fallbackStore is provided', () => {
+      cacheConfig.USE_REDIS = false;
+      const namespace = 'test-namespace';
+      const ttl = 3600;
+      const fallbackStore = { some: 'store' };
+
+      standardCache(namespace, ttl, fallbackStore);
+
+      expect(mockKeyv).toHaveBeenCalledWith({ store: fallbackStore, namespace, ttl });
+    });
+
+    it('should create default Keyv instance when USE_REDIS is false and no fallbackStore', () => {
+      cacheConfig.USE_REDIS = false;
+      const namespace = 'test-namespace';
+      const ttl = 3600;
+
+      standardCache(namespace, ttl);
+
+      expect(mockKeyv).toHaveBeenCalledWith({ namespace, ttl });
+    });
+
+    it('should handle namespace and ttl as undefined', () => {
+      cacheConfig.USE_REDIS = false;
+
+      standardCache();
+
+      expect(mockKeyv).toHaveBeenCalledWith({ namespace: undefined, ttl: undefined });
+    });
+
+    it('should use fallback when namespace is in FORCED_IN_MEMORY_CACHE_NAMESPACES', () => {
+      cacheConfig.USE_REDIS = true;
+      cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES = ['forced-memory'];
+      const namespace = 'forced-memory';
+      const ttl = 3600;
+
+      standardCache(namespace, ttl);
+
+      expect(require('@keyv/redis').default).not.toHaveBeenCalled();
+      expect(mockKeyv).toHaveBeenCalledWith({ namespace, ttl });
+    });
+
+    it('should use Redis when namespace is not in FORCED_IN_MEMORY_CACHE_NAMESPACES', () => {
+      cacheConfig.USE_REDIS = true;
+      cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES = ['other-namespace'];
+      const namespace = 'test-namespace';
+      const ttl = 3600;
+
+      standardCache(namespace, ttl);
+
+      expect(require('@keyv/redis').default).toHaveBeenCalledWith(mockKeyvRedisClient);
+      expect(mockKeyv).toHaveBeenCalledWith(mockKeyvRedis, { namespace, ttl });
+    });
+
+    it('should throw error when Redis cache creation fails', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'test-namespace';
+      const ttl = 3600;
+      const testError = new Error('Redis connection failed');
+
+      const KeyvRedis = require('@keyv/redis').default;
+      KeyvRedis.mockImplementationOnce(() => {
+        throw testError;
+      });
+
+      expect(() => standardCache(namespace, ttl)).toThrow('Redis connection failed');
+
+      const { logger } = require('@librechat/data-schemas');
+      expect(logger.error).toHaveBeenCalledWith(
+        `Failed to create Redis cache for namespace ${namespace}:`,
+        testError,
+      );
+
+      expect(mockKeyv).not.toHaveBeenCalled();
+    });
+  });
+
+  describe('violationCache', () => {
+    it('should create violation cache with prefixed namespace', () => {
+      const namespace = 'test-violations';
+      const ttl = 7200;
+
+      // We can't easily mock the internal redisCache call since it's in the same module
+      // But we can test that the function executes without throwing
+      expect(() => violationCache(namespace, ttl)).not.toThrow();
+    });
+
+    it('should create violation cache with undefined ttl', () => {
+      const namespace = 'test-violations';
+
+      violationCache(namespace);
+
+      // The function should call redisCache with violations: prefixed namespace
+      // Since we can't easily mock the internal redisCache call, we test the behavior
+      expect(() => violationCache(namespace)).not.toThrow();
+    });
+
+    it('should handle undefined namespace', () => {
+      expect(() => violationCache(undefined)).not.toThrow();
+    });
+  });
+
+  describe('sessionCache', () => {
+    it('should return MemoryStore when USE_REDIS is false', () => {
+      cacheConfig.USE_REDIS = false;
+      const namespace = 'sessions';
+      const ttl = 86400;
+
+      const result = sessionCache(namespace, ttl);
+
+      expect(mockMemoryStore).toHaveBeenCalledWith({ ttl, checkPeriod: Time.ONE_DAY });
+      expect(result).toBe(mockMemoryStore());
+    });
+
+    it('should return ConnectRedis when USE_REDIS is true', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions';
+      const ttl = 86400;
+
+      const result = sessionCache(namespace, ttl);
+
+      expect(mockConnectRedis).toHaveBeenCalledWith({
+        client: mockIoredisClient,
+        ttl,
+        prefix: `${namespace}:`,
+      });
+      expect(result).toBe(mockConnectRedis());
+    });
+
+    it('should add colon to namespace if not present', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions';
+
+      sessionCache(namespace);
+
+      expect(mockConnectRedis).toHaveBeenCalledWith({
+        client: mockIoredisClient,
+        ttl: undefined,
+        prefix: 'sessions:',
+      });
+    });
+
+    it('should not add colon to namespace if already present', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions:';
+
+      sessionCache(namespace);
+
+      expect(mockConnectRedis).toHaveBeenCalledWith({
+        client: mockIoredisClient,
+        ttl: undefined,
+        prefix: 'sessions:',
+      });
+    });
+
+    it('should handle undefined ttl', () => {
+      cacheConfig.USE_REDIS = false;
+      const namespace = 'sessions';
+
+      sessionCache(namespace);
+
+      expect(mockMemoryStore).toHaveBeenCalledWith({
+        ttl: undefined,
+        checkPeriod: Time.ONE_DAY,
+      });
+    });
+
+    it('should throw error when ConnectRedis constructor fails', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions';
+      const ttl = 86400;
+
+      // Mock ConnectRedis to throw an error during construction
+      const redisError = new Error('Redis connection failed');
+      mockConnectRedis.mockImplementationOnce(() => {
+        throw redisError;
+      });
+
+      // The error should propagate up, not be caught
+      expect(() => sessionCache(namespace, ttl)).toThrow('Redis connection failed');
+
+      // Verify that MemoryStore was NOT used as fallback
+      expect(mockMemoryStore).not.toHaveBeenCalled();
+    });
+
+    it('should register error handler but let errors propagate to Express', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions';
+
+      // Create a mock session store with middleware methods
+      const mockSessionStore = {
+        get: jest.fn(),
+        set: jest.fn(),
+        destroy: jest.fn(),
+      };
+      mockConnectRedis.mockReturnValue(mockSessionStore);
+
+      const store = sessionCache(namespace);
+
+      // Verify error handler was registered
+      expect(mockIoredisClient.on).toHaveBeenCalledWith('error', expect.any(Function));
+
+      // Get the error handler
+      const errorHandler = mockIoredisClient.on.mock.calls.find((call) => call[0] === 'error')[1];
+
+      // Simulate an error from Redis during a session operation
+      const redisError = new Error('Socket closed unexpectedly');
+
+      // The error handler should log but not swallow the error
+      const { logger } = require('@librechat/data-schemas');
+      errorHandler(redisError);
+
+      expect(logger.error).toHaveBeenCalledWith(
+        `Session store Redis error for namespace ${namespace}::`,
+        redisError,
+      );
+
+      // Now simulate what happens when session middleware tries to use the store
+      const callback = jest.fn();
+      mockSessionStore.get.mockImplementation((sid, cb) => {
+        cb(new Error('Redis connection lost'));
+      });
+
+      // Call the store's get method (as Express session would)
+      store.get('test-session-id', callback);
+
+      // The error should be passed to the callback, not swallowed
+      expect(callback).toHaveBeenCalledWith(new Error('Redis connection lost'));
+    });
+
+    it('should handle null ioredisClient gracefully', () => {
+      cacheConfig.USE_REDIS = true;
+      const namespace = 'sessions';
+
+      // Temporarily set ioredisClient to null (simulating connection not established)
+      const originalClient = require('./redisClients').ioredisClient;
+      require('./redisClients').ioredisClient = null;
+
+      // ConnectRedis might accept null client but would fail on first use
+      // The important thing is it doesn't throw uncaught exceptions during construction
+      const store = sessionCache(namespace);
+      expect(store).toBeDefined();
+
+      // Restore original client
+      require('./redisClients').ioredisClient = originalClient;
+    });
+  });
+
+  describe('limiterCache', () => {
+    it('should return undefined when USE_REDIS is false', () => {
+      cacheConfig.USE_REDIS = false;
+      const result = limiterCache('prefix');
+
+      expect(result).toBeUndefined();
+    });
+
+    it('should return RedisStore when USE_REDIS is true', () => {
+      cacheConfig.USE_REDIS = true;
+      const result = limiterCache('rate-limit');
+
+      expect(mockRedisStore).toHaveBeenCalledWith({
+        sendCommand: expect.any(Function),
+        prefix: `rate-limit:`,
+      });
+      expect(result).toBe(mockRedisStore());
+    });
+
+    it('should add colon to prefix if not present', () => {
+      cacheConfig.USE_REDIS = true;
+      limiterCache('rate-limit');
+
+      expect(mockRedisStore).toHaveBeenCalledWith({
+        sendCommand: expect.any(Function),
+        prefix: 'rate-limit:',
+      });
+    });
+
+    it('should not add colon to prefix if already present', () => {
+      cacheConfig.USE_REDIS = true;
+      limiterCache('rate-limit:');
+
+      expect(mockRedisStore).toHaveBeenCalledWith({
+        sendCommand: expect.any(Function),
+        prefix: 'rate-limit:',
+      });
+    });
+
+    it('should pass sendCommand function that calls ioredisClient.call', async () => {
+      cacheConfig.USE_REDIS = true;
+      mockIoredisClient.call.mockResolvedValue('test-value');
+
+      limiterCache('rate-limit');
+
+      const sendCommandCall = mockRedisStore.mock.calls[0][0];
+      const sendCommand = sendCommandCall.sendCommand;
+
+      // Test that sendCommand properly delegates to ioredisClient.call
+      const args = ['GET', 'test-key'];
+      const result = await sendCommand(...args);
+
+      expect(mockIoredisClient.call).toHaveBeenCalledWith(...args);
+      expect(result).toBe('test-value');
+    });
+
+    it('should handle sendCommand errors properly', async () => {
+      cacheConfig.USE_REDIS = true;
+
+      // Mock the call method to reject with an error
+      const testError = new Error('Redis error');
+      mockIoredisClient.call.mockRejectedValue(testError);
+
+      limiterCache('rate-limit');
+
+      const sendCommandCall = mockRedisStore.mock.calls[0][0];
+      const sendCommand = sendCommandCall.sendCommand;
+
+      // Test that sendCommand properly handles errors
+      const args = ['GET', 'test-key'];
+
+      await expect(sendCommand(...args)).rejects.toThrow('Redis error');
+      expect(mockIoredisClient.call).toHaveBeenCalledWith(...args);
+    });
+
+    it('should handle undefined prefix', () => {
+      cacheConfig.USE_REDIS = true;
+      expect(() => limiterCache()).toThrow('prefix is required');
+    });
+  });
+});

api/cache/getLogStores.js

@@ -1,108 +1,52 @@
+const { cacheConfig } = require('./cacheConfig');
 const { Keyv } = require('keyv');
 const { CacheKeys, ViolationTypes, Time } = require('librechat-data-provider');
-const { logFile, violationFile } = require('./keyvFiles');
-const { isEnabled, math } = require('~/server/utils');
-const keyvRedis = require('./keyvRedis');
+const { logFile } = require('./keyvFiles');
 const keyvMongo = require('./keyvMongo');
-
-const { BAN_DURATION, USE_REDIS, DEBUG_MEMORY_CACHE, CI } = process.env ?? {};
-
-const duration = math(BAN_DURATION, 7200000);
-const isRedisEnabled = isEnabled(USE_REDIS);
-const debugMemoryCache = isEnabled(DEBUG_MEMORY_CACHE);
-
-const createViolationInstance = (namespace) => {
-  const config = isRedisEnabled ? { store: keyvRedis } : { store: violationFile, namespace };
-  return new Keyv(config);
-};
-
-// Serve cache from memory so no need to clear it on startup/exit
-const pending_req = isRedisEnabled
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.PENDING_REQ });
-
-const config = isRedisEnabled
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.CONFIG_STORE });
-
-const roles = isRedisEnabled
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.ROLES });
-
-const audioRuns = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.TEN_MINUTES })
-  : new Keyv({ namespace: CacheKeys.AUDIO_RUNS, ttl: Time.TEN_MINUTES });
-
-const messages = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.ONE_MINUTE })
-  : new Keyv({ namespace: CacheKeys.MESSAGES, ttl: Time.ONE_MINUTE });
-
-const flows = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.TWO_MINUTES })
-  : new Keyv({ namespace: CacheKeys.FLOWS, ttl: Time.ONE_MINUTE * 3 });
-
-const tokenConfig = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.THIRTY_MINUTES })
-  : new Keyv({ namespace: CacheKeys.TOKEN_CONFIG, ttl: Time.THIRTY_MINUTES });
-
-const genTitle = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.TWO_MINUTES })
-  : new Keyv({ namespace: CacheKeys.GEN_TITLE, ttl: Time.TWO_MINUTES });
-
-const s3ExpiryInterval = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.THIRTY_MINUTES })
-  : new Keyv({ namespace: CacheKeys.S3_EXPIRY_INTERVAL, ttl: Time.THIRTY_MINUTES });
-
-const modelQueries = isEnabled(process.env.USE_REDIS)
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.MODEL_QUERIES });
-
-const abortKeys = isRedisEnabled
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.ABORT_KEYS, ttl: Time.TEN_MINUTES });
-
-const openIdExchangedTokensCache = isRedisEnabled
-  ? new Keyv({ store: keyvRedis, ttl: Time.TEN_MINUTES })
-  : new Keyv({ namespace: CacheKeys.OPENID_EXCHANGED_TOKENS, ttl: Time.TEN_MINUTES });
+const { standardCache, sessionCache, violationCache } = require('./cacheFactory');
 
 const namespaces = {
-  [CacheKeys.ROLES]: roles,
-  [CacheKeys.CONFIG_STORE]: config,
-  [CacheKeys.PENDING_REQ]: pending_req,
-  [ViolationTypes.BAN]: new Keyv({ store: keyvMongo, namespace: CacheKeys.BANS, ttl: duration }),
-  [CacheKeys.ENCODED_DOMAINS]: new Keyv({
+  [ViolationTypes.GENERAL]: new Keyv({ store: logFile, namespace: 'violations' }),
+  [ViolationTypes.LOGINS]: violationCache(ViolationTypes.LOGINS),
+  [ViolationTypes.CONCURRENT]: violationCache(ViolationTypes.CONCURRENT),
+  [ViolationTypes.NON_BROWSER]: violationCache(ViolationTypes.NON_BROWSER),
+  [ViolationTypes.MESSAGE_LIMIT]: violationCache(ViolationTypes.MESSAGE_LIMIT),
+  [ViolationTypes.REGISTRATIONS]: violationCache(ViolationTypes.REGISTRATIONS),
+  [ViolationTypes.TOKEN_BALANCE]: violationCache(ViolationTypes.TOKEN_BALANCE),
+  [ViolationTypes.TTS_LIMIT]: violationCache(ViolationTypes.TTS_LIMIT),
+  [ViolationTypes.STT_LIMIT]: violationCache(ViolationTypes.STT_LIMIT),
+  [ViolationTypes.CONVO_ACCESS]: violationCache(ViolationTypes.CONVO_ACCESS),
+  [ViolationTypes.TOOL_CALL_LIMIT]: violationCache(ViolationTypes.TOOL_CALL_LIMIT),
+  [ViolationTypes.FILE_UPLOAD_LIMIT]: violationCache(ViolationTypes.FILE_UPLOAD_LIMIT),
+  [ViolationTypes.VERIFY_EMAIL_LIMIT]: violationCache(ViolationTypes.VERIFY_EMAIL_LIMIT),
+  [ViolationTypes.RESET_PASSWORD_LIMIT]: violationCache(ViolationTypes.RESET_PASSWORD_LIMIT),
+  [ViolationTypes.ILLEGAL_MODEL_REQUEST]: violationCache(ViolationTypes.ILLEGAL_MODEL_REQUEST),
+  [ViolationTypes.BAN]: new Keyv({
     store: keyvMongo,
-    namespace: CacheKeys.ENCODED_DOMAINS,
-    ttl: 0,
+    namespace: CacheKeys.BANS,
+    ttl: cacheConfig.BAN_DURATION,
   }),
-  general: new Keyv({ store: logFile, namespace: 'violations' }),
-  concurrent: createViolationInstance('concurrent'),
-  non_browser: createViolationInstance('non_browser'),
-  message_limit: createViolationInstance('message_limit'),
-  token_balance: createViolationInstance(ViolationTypes.TOKEN_BALANCE),
-  registrations: createViolationInstance('registrations'),
-  [ViolationTypes.TTS_LIMIT]: createViolationInstance(ViolationTypes.TTS_LIMIT),
-  [ViolationTypes.STT_LIMIT]: createViolationInstance(ViolationTypes.STT_LIMIT),
-  [ViolationTypes.CONVO_ACCESS]: createViolationInstance(ViolationTypes.CONVO_ACCESS),
-  [ViolationTypes.TOOL_CALL_LIMIT]: createViolationInstance(ViolationTypes.TOOL_CALL_LIMIT),
-  [ViolationTypes.FILE_UPLOAD_LIMIT]: createViolationInstance(ViolationTypes.FILE_UPLOAD_LIMIT),
-  [ViolationTypes.VERIFY_EMAIL_LIMIT]: createViolationInstance(ViolationTypes.VERIFY_EMAIL_LIMIT),
-  [ViolationTypes.RESET_PASSWORD_LIMIT]: createViolationInstance(
-    ViolationTypes.RESET_PASSWORD_LIMIT,
+
+  [CacheKeys.OPENID_SESSION]: sessionCache(CacheKeys.OPENID_SESSION),
+  [CacheKeys.SAML_SESSION]: sessionCache(CacheKeys.SAML_SESSION),
+
+  [CacheKeys.ROLES]: standardCache(CacheKeys.ROLES),
+  [CacheKeys.APP_CONFIG]: standardCache(CacheKeys.APP_CONFIG),
+  [CacheKeys.CONFIG_STORE]: standardCache(CacheKeys.CONFIG_STORE),
+  [CacheKeys.PENDING_REQ]: standardCache(CacheKeys.PENDING_REQ),
+  [CacheKeys.ENCODED_DOMAINS]: new Keyv({ store: keyvMongo, namespace: CacheKeys.ENCODED_DOMAINS }),
+  [CacheKeys.ABORT_KEYS]: standardCache(CacheKeys.ABORT_KEYS, Time.TEN_MINUTES),
+  [CacheKeys.TOKEN_CONFIG]: standardCache(CacheKeys.TOKEN_CONFIG, Time.THIRTY_MINUTES),
+  [CacheKeys.GEN_TITLE]: standardCache(CacheKeys.GEN_TITLE, Time.TWO_MINUTES),
+  [CacheKeys.S3_EXPIRY_INTERVAL]: standardCache(CacheKeys.S3_EXPIRY_INTERVAL, Time.THIRTY_MINUTES),
+  [CacheKeys.MODEL_QUERIES]: standardCache(CacheKeys.MODEL_QUERIES),
+  [CacheKeys.AUDIO_RUNS]: standardCache(CacheKeys.AUDIO_RUNS, Time.TEN_MINUTES),
+  [CacheKeys.MESSAGES]: standardCache(CacheKeys.MESSAGES, Time.ONE_MINUTE),
+  [CacheKeys.FLOWS]: standardCache(CacheKeys.FLOWS, Time.ONE_MINUTE * 3),
+  [CacheKeys.OPENID_EXCHANGED_TOKENS]: standardCache(
+    CacheKeys.OPENID_EXCHANGED_TOKENS,
+    Time.TEN_MINUTES,
   ),
-  [ViolationTypes.ILLEGAL_MODEL_REQUEST]: createViolationInstance(
-    ViolationTypes.ILLEGAL_MODEL_REQUEST,
-  ),
-  logins: createViolationInstance('logins'),
-  [CacheKeys.ABORT_KEYS]: abortKeys,
-  [CacheKeys.TOKEN_CONFIG]: tokenConfig,
-  [CacheKeys.GEN_TITLE]: genTitle,
-  [CacheKeys.S3_EXPIRY_INTERVAL]: s3ExpiryInterval,
-  [CacheKeys.MODEL_QUERIES]: modelQueries,
-  [CacheKeys.AUDIO_RUNS]: audioRuns,
-  [CacheKeys.MESSAGES]: messages,
-  [CacheKeys.FLOWS]: flows,
-  [CacheKeys.OPENID_EXCHANGED_TOKENS]: openIdExchangedTokensCache,
 };
 
 /**
@@ -111,7 +55,10 @@ const namespaces = {
  */
 function getTTLStores() {
   return Object.values(namespaces).filter(
-    (store) => store instanceof Keyv && typeof store.opts?.ttl === 'number' && store.opts.ttl > 0,
+    (store) =>
+      store instanceof Keyv &&
+      parseInt(store.opts?.ttl ?? '0') > 0 &&
+      !store.opts?.store?.constructor?.name?.includes('Redis'), // Only include non-Redis stores
   );
 }
 
@@ -147,18 +94,18 @@ async function clearExpiredFromCache(cache) {
       if (data?.expires && data.expires <= expiryTime) {
         const deleted = await cache.opts.store.delete(key);
         if (!deleted) {
-          debugMemoryCache &&
+          cacheConfig.DEBUG_MEMORY_CACHE &&
             console.warn(`[Cache] Error deleting entry: ${key} from ${cache.opts.namespace}`);
           continue;
         }
         cleared++;
       }
     } catch (error) {
-      debugMemoryCache &&
+      cacheConfig.DEBUG_MEMORY_CACHE &&
         console.log(`[Cache] Error processing entry from ${cache.opts.namespace}:`, error);
       const deleted = await cache.opts.store.delete(key);
       if (!deleted) {
-        debugMemoryCache &&
+        cacheConfig.DEBUG_MEMORY_CACHE &&
           console.warn(`[Cache] Error deleting entry: ${key} from ${cache.opts.namespace}`);
         continue;
       }
@@ -167,7 +114,7 @@ async function clearExpiredFromCache(cache) {
   }
 
   if (cleared > 0) {
-    debugMemoryCache &&
+    cacheConfig.DEBUG_MEMORY_CACHE &&
       console.log(
         `[Cache] Cleared ${cleared} entries older than ${ttl}ms from ${cache.opts.namespace}`,
       );
@@ -208,7 +155,7 @@ async function clearAllExpiredFromCache() {
   }
 }
 
-if (!isRedisEnabled && !isEnabled(CI)) {
+if (!cacheConfig.USE_REDIS && !cacheConfig.CI) {
   /** @type {Set<NodeJS.Timeout>} */
   const cleanupIntervals = new Set();
 
@@ -219,7 +166,7 @@ if (!isRedisEnabled && !isEnabled(CI)) {
 
   cleanupIntervals.add(cleanup);
 
-  if (debugMemoryCache) {
+  if (cacheConfig.DEBUG_MEMORY_CACHE) {
     const monitor = setInterval(() => {
       const ttlStores = getTTLStores();
       const memory = process.memoryUsage();
@@ -240,13 +187,13 @@ if (!isRedisEnabled && !isEnabled(CI)) {
   }
 
   const dispose = () => {
-    debugMemoryCache && console.log('[Cache] Cleaning up and shutting down...');
+    cacheConfig.DEBUG_MEMORY_CACHE && console.log('[Cache] Cleaning up and shutting down...');
     cleanupIntervals.forEach((interval) => clearInterval(interval));
     cleanupIntervals.clear();
 
     // One final cleanup before exit
     clearAllExpiredFromCache().then(() => {
-      debugMemoryCache && console.log('[Cache] Final cleanup completed');
+      cacheConfig.DEBUG_MEMORY_CACHE && console.log('[Cache] Final cleanup completed');
       process.exit(0);
     });
   };

api/cache/ioredisClient.js

@@ -1,92 +0,0 @@
-const fs = require('fs');
-const Redis = require('ioredis');
-const { isEnabled } = require('~/server/utils');
-const logger = require('~/config/winston');
-
-const { REDIS_URI, USE_REDIS, USE_REDIS_CLUSTER, REDIS_CA, REDIS_MAX_LISTENERS } = process.env;
-
-/** @type {import('ioredis').Redis | import('ioredis').Cluster} */
-let ioredisClient;
-const redis_max_listeners = Number(REDIS_MAX_LISTENERS) || 40;
-
-function mapURI(uri) {
-  const regex =
-    /^(?:(?<scheme>\w+):\/\/)?(?:(?<user>[^:@]+)(?::(?<password>[^@]+))?@)?(?<host>[\w.-]+)(?::(?<port>\d{1,5}))?$/;
-  const match = uri.match(regex);
-
-  if (match) {
-    const { scheme, user, password, host, port } = match.groups;
-
-    return {
-      scheme: scheme || 'none',
-      user: user || null,
-      password: password || null,
-      host: host || null,
-      port: port || null,
-    };
-  } else {
-    const parts = uri.split(':');
-    if (parts.length === 2) {
-      return {
-        scheme: 'none',
-        user: null,
-        password: null,
-        host: parts[0],
-        port: parts[1],
-      };
-    }
-
-    return {
-      scheme: 'none',
-      user: null,
-      password: null,
-      host: uri,
-      port: null,
-    };
-  }
-}
-
-if (REDIS_URI && isEnabled(USE_REDIS)) {
-  let redisOptions = null;
-
-  if (REDIS_CA) {
-    const ca = fs.readFileSync(REDIS_CA);
-    redisOptions = { tls: { ca } };
-  }
-
-  if (isEnabled(USE_REDIS_CLUSTER)) {
-    const hosts = REDIS_URI.split(',').map((item) => {
-      var value = mapURI(item);
-
-      return {
-        host: value.host,
-        port: value.port,
-      };
-    });
-    ioredisClient = new Redis.Cluster(hosts, { redisOptions });
-  } else {
-    ioredisClient = new Redis(REDIS_URI, redisOptions);
-  }
-
-  ioredisClient.on('ready', () => {
-    logger.info('IoRedis connection ready');
-  });
-  ioredisClient.on('reconnecting', () => {
-    logger.info('IoRedis connection reconnecting');
-  });
-  ioredisClient.on('end', () => {
-    logger.info('IoRedis connection ended');
-  });
-  ioredisClient.on('close', () => {
-    logger.info('IoRedis connection closed');
-  });
-  ioredisClient.on('error', (err) => logger.error('IoRedis connection error:', err));
-  ioredisClient.setMaxListeners(redis_max_listeners);
-  logger.info(
-    '[Optional] IoRedis initialized for rate limiters. If you have issues, disable Redis or restart the server.',
-  );
-} else {
-  logger.info('[Optional] IoRedis not initialized for rate limiters.');
-}
-
-module.exports = ioredisClient;

api/cache/keyvRedis.js

@@ -1,109 +0,0 @@
-const fs = require('fs');
-const ioredis = require('ioredis');
-const KeyvRedis = require('@keyv/redis').default;
-const { isEnabled } = require('~/server/utils');
-const logger = require('~/config/winston');
-
-const { REDIS_URI, USE_REDIS, USE_REDIS_CLUSTER, REDIS_CA, REDIS_KEY_PREFIX, REDIS_MAX_LISTENERS } =
-  process.env;
-
-let keyvRedis;
-const redis_prefix = REDIS_KEY_PREFIX || '';
-const redis_max_listeners = Number(REDIS_MAX_LISTENERS) || 40;
-
-function mapURI(uri) {
-  const regex =
-    /^(?:(?<scheme>\w+):\/\/)?(?:(?<user>[^:@]+)(?::(?<password>[^@]+))?@)?(?<host>[\w.-]+)(?::(?<port>\d{1,5}))?$/;
-  const match = uri.match(regex);
-
-  if (match) {
-    const { scheme, user, password, host, port } = match.groups;
-
-    return {
-      scheme: scheme || 'none',
-      user: user || null,
-      password: password || null,
-      host: host || null,
-      port: port || null,
-    };
-  } else {
-    const parts = uri.split(':');
-    if (parts.length === 2) {
-      return {
-        scheme: 'none',
-        user: null,
-        password: null,
-        host: parts[0],
-        port: parts[1],
-      };
-    }
-
-    return {
-      scheme: 'none',
-      user: null,
-      password: null,
-      host: uri,
-      port: null,
-    };
-  }
-}
-
-if (REDIS_URI && isEnabled(USE_REDIS)) {
-  let redisOptions = null;
-  /** @type {import('@keyv/redis').KeyvRedisOptions} */
-  let keyvOpts = {
-    useRedisSets: false,
-    keyPrefix: redis_prefix,
-  };
-
-  if (REDIS_CA) {
-    const ca = fs.readFileSync(REDIS_CA);
-    redisOptions = { tls: { ca } };
-  }
-
-  if (isEnabled(USE_REDIS_CLUSTER)) {
-    const hosts = REDIS_URI.split(',').map((item) => {
-      var value = mapURI(item);
-
-      return {
-        host: value.host,
-        port: value.port,
-      };
-    });
-    const cluster = new ioredis.Cluster(hosts, { redisOptions });
-    keyvRedis = new KeyvRedis(cluster, keyvOpts);
-  } else {
-    keyvRedis = new KeyvRedis(REDIS_URI, keyvOpts);
-  }
-
-  const pingInterval = setInterval(
-    () => {
-      logger.debug('KeyvRedis ping');
-      keyvRedis.client.ping().catch((err) => logger.error('Redis keep-alive ping failed:', err));
-    },
-    5 * 60 * 1000,
-  );
-
-  keyvRedis.on('ready', () => {
-    logger.info('KeyvRedis connection ready');
-  });
-  keyvRedis.on('reconnecting', () => {
-    logger.info('KeyvRedis connection reconnecting');
-  });
-  keyvRedis.on('end', () => {
-    logger.info('KeyvRedis connection ended');
-  });
-  keyvRedis.on('close', () => {
-    clearInterval(pingInterval);
-    logger.info('KeyvRedis connection closed');
-  });
-  keyvRedis.on('error', (err) => logger.error('KeyvRedis connection error:', err));
-  keyvRedis.setMaxListeners(redis_max_listeners);
-  logger.info(
-    '[Optional] Redis initialized. If you have issues, or seeing older values, disable it or flush cache to refresh values.',
-  );
-} else {
-  logger.info('[Optional] Redis not initialized.');
-}
-
-module.exports = keyvRedis;

api/cache/logViolation.js

@@ -1,4 +1,5 @@
 const { isEnabled } = require('~/server/utils');
+const { ViolationTypes } = require('librechat-data-provider');
 const getLogStores = require('./getLogStores');
 const banViolation = require('./banViolation');
 
@@ -9,14 +10,14 @@ const banViolation = require('./banViolation');
  * @param {Object} res - Express response object.
  * @param {string} type - The type of violation.
  * @param {Object} errorMessage - The error message to log.
- * @param {number} [score=1] - The severity of the violation. Defaults to 1
+ * @param {number | string} [score=1] - The severity of the violation. Defaults to 1
  */
 const logViolation = async (req, res, type, errorMessage, score = 1) => {
   const userId = req.user?.id ?? req.user?._id;
   if (!userId) {
     return;
   }
-  const logs = getLogStores('general');
+  const logs = getLogStores(ViolationTypes.GENERAL);
   const violationLogs = getLogStores(type);
   const key = isEnabled(process.env.USE_REDIS) ? `${type}:${userId}` : userId;
 

api/cache/redisClients.js

@@ -0,0 +1,210 @@
+const IoRedis = require('ioredis');
+const { logger } = require('@librechat/data-schemas');
+const { createClient, createCluster } = require('@keyv/redis');
+const { cacheConfig } = require('./cacheConfig');
+
+const GLOBAL_PREFIX_SEPARATOR = '::';
+
+const urls = cacheConfig.REDIS_URI?.split(',').map((uri) => new URL(uri));
+const username = urls?.[0].username || cacheConfig.REDIS_USERNAME;
+const password = urls?.[0].password || cacheConfig.REDIS_PASSWORD;
+const ca = cacheConfig.REDIS_CA;
+
+/** @type {import('ioredis').Redis | import('ioredis').Cluster | null} */
+let ioredisClient = null;
+if (cacheConfig.USE_REDIS) {
+  /** @type {import('ioredis').RedisOptions | import('ioredis').ClusterOptions} */
+  const redisOptions = {
+    username: username,
+    password: password,
+    tls: ca ? { ca } : undefined,
+    keyPrefix: `${cacheConfig.REDIS_KEY_PREFIX}${GLOBAL_PREFIX_SEPARATOR}`,
+    maxListeners: cacheConfig.REDIS_MAX_LISTENERS,
+    retryStrategy: (times) => {
+      if (
+        cacheConfig.REDIS_RETRY_MAX_ATTEMPTS > 0 &&
+        times > cacheConfig.REDIS_RETRY_MAX_ATTEMPTS
+      ) {
+        logger.error(
+          `ioredis giving up after ${cacheConfig.REDIS_RETRY_MAX_ATTEMPTS} reconnection attempts`,
+        );
+        return null;
+      }
+      const delay = Math.min(times * 50, cacheConfig.REDIS_RETRY_MAX_DELAY);
+      logger.info(`ioredis reconnecting... attempt ${times}, delay ${delay}ms`);
+      return delay;
+    },
+    reconnectOnError: (err) => {
+      const targetError = 'READONLY';
+      if (err.message.includes(targetError)) {
+        logger.warn('ioredis reconnecting due to READONLY error');
+        return 2; // Return retry delay instead of boolean
+      }
+      return false;
+    },
+    enableOfflineQueue: cacheConfig.REDIS_ENABLE_OFFLINE_QUEUE,
+    connectTimeout: cacheConfig.REDIS_CONNECT_TIMEOUT,
+    maxRetriesPerRequest: 3,
+  };
+
+  ioredisClient =
+    urls.length === 1 && !cacheConfig.USE_REDIS_CLUSTER
+      ? new IoRedis(cacheConfig.REDIS_URI, redisOptions)
+      : new IoRedis.Cluster(
+          urls.map((url) => ({ host: url.hostname, port: parseInt(url.port, 10) || 6379 })),
+          {
+            ...(cacheConfig.REDIS_USE_ALTERNATIVE_DNS_LOOKUP
+              ? { dnsLookup: (address, callback) => callback(null, address) }
+              : {}),
+            redisOptions,
+            clusterRetryStrategy: (times) => {
+              if (
+                cacheConfig.REDIS_RETRY_MAX_ATTEMPTS > 0 &&
+                times > cacheConfig.REDIS_RETRY_MAX_ATTEMPTS
+              ) {
+                logger.error(
+                  `ioredis cluster giving up after ${cacheConfig.REDIS_RETRY_MAX_ATTEMPTS} reconnection attempts`,
+                );
+                return null;
+              }
+              const delay = Math.min(times * 100, cacheConfig.REDIS_RETRY_MAX_DELAY);
+              logger.info(`ioredis cluster reconnecting... attempt ${times}, delay ${delay}ms`);
+              return delay;
+            },
+            enableOfflineQueue: cacheConfig.REDIS_ENABLE_OFFLINE_QUEUE,
+          },
+        );
+
+  ioredisClient.on('error', (err) => {
+    logger.error('ioredis client error:', err);
+  });
+
+  ioredisClient.on('connect', () => {
+    logger.info('ioredis client connected');
+  });
+
+  ioredisClient.on('ready', () => {
+    logger.info('ioredis client ready');
+  });
+
+  ioredisClient.on('reconnecting', (delay) => {
+    logger.info(`ioredis client reconnecting in ${delay}ms`);
+  });
+
+  ioredisClient.on('close', () => {
+    logger.warn('ioredis client connection closed');
+  });
+
+  /** Ping Interval to keep the Redis server connection alive (if enabled) */
+  let pingInterval = null;
+  const clearPingInterval = () => {
+    if (pingInterval) {
+      clearInterval(pingInterval);
+      pingInterval = null;
+    }
+  };
+
+  if (cacheConfig.REDIS_PING_INTERVAL > 0) {
+    pingInterval = setInterval(() => {
+      if (ioredisClient && ioredisClient.status === 'ready') {
+        ioredisClient.ping().catch((err) => {
+          logger.error('ioredis ping failed:', err);
+        });
+      }
+    }, cacheConfig.REDIS_PING_INTERVAL * 1000);
+    ioredisClient.on('close', clearPingInterval);
+    ioredisClient.on('end', clearPingInterval);
+  }
+}
+
+/** @type {import('@keyv/redis').RedisClient | import('@keyv/redis').RedisCluster | null} */
+let keyvRedisClient = null;
+if (cacheConfig.USE_REDIS) {
+  /**
+   * ** WARNING ** Keyv Redis client does not support Prefix like ioredis above.
+   * The prefix feature will be handled by the Keyv-Redis store in cacheFactory.js
+   * @type {import('@keyv/redis').RedisClientOptions | import('@keyv/redis').RedisClusterOptions}
+   */
+  const redisOptions = {
+    username,
+    password,
+    socket: {
+      tls: ca != null,
+      ca,
+      connectTimeout: cacheConfig.REDIS_CONNECT_TIMEOUT,
+      reconnectStrategy: (retries) => {
+        if (
+          cacheConfig.REDIS_RETRY_MAX_ATTEMPTS > 0 &&
+          retries > cacheConfig.REDIS_RETRY_MAX_ATTEMPTS
+        ) {
+          logger.error(
+            `@keyv/redis client giving up after ${cacheConfig.REDIS_RETRY_MAX_ATTEMPTS} reconnection attempts`,
+          );
+          return new Error('Max reconnection attempts reached');
+        }
+        const delay = Math.min(retries * 100, cacheConfig.REDIS_RETRY_MAX_DELAY);
+        logger.info(`@keyv/redis reconnecting... attempt ${retries}, delay ${delay}ms`);
+        return delay;
+      },
+    },
+    disableOfflineQueue: !cacheConfig.REDIS_ENABLE_OFFLINE_QUEUE,
+  };
+
+  keyvRedisClient =
+    urls.length === 1 && !cacheConfig.USE_REDIS_CLUSTER
+      ? createClient({ url: cacheConfig.REDIS_URI, ...redisOptions })
+      : createCluster({
+          rootNodes: urls.map((url) => ({ url: url.href })),
+          defaults: redisOptions,
+        });
+
+  keyvRedisClient.setMaxListeners(cacheConfig.REDIS_MAX_LISTENERS);
+
+  keyvRedisClient.on('error', (err) => {
+    logger.error('@keyv/redis client error:', err);
+  });
+
+  keyvRedisClient.on('connect', () => {
+    logger.info('@keyv/redis client connected');
+  });
+
+  keyvRedisClient.on('ready', () => {
+    logger.info('@keyv/redis client ready');
+  });
+
+  keyvRedisClient.on('reconnecting', () => {
+    logger.info('@keyv/redis client reconnecting...');
+  });
+
+  keyvRedisClient.on('disconnect', () => {
+    logger.warn('@keyv/redis client disconnected');
+  });
+
+  keyvRedisClient.connect().catch((err) => {
+    logger.error('@keyv/redis initial connection failed:', err);
+    throw err;
+  });
+
+  /** Ping Interval to keep the Redis server connection alive (if enabled) */
+  let pingInterval = null;
+  const clearPingInterval = () => {
+    if (pingInterval) {
+      clearInterval(pingInterval);
+      pingInterval = null;
+    }
+  };
+
+  if (cacheConfig.REDIS_PING_INTERVAL > 0) {
+    pingInterval = setInterval(() => {
+      if (keyvRedisClient && keyvRedisClient.isReady) {
+        keyvRedisClient.ping().catch((err) => {
+          logger.error('@keyv/redis ping failed:', err);
+        });
+      }
+    }, cacheConfig.REDIS_PING_INTERVAL * 1000);
+    keyvRedisClient.on('disconnect', clearPingInterval);
+    keyvRedisClient.on('end', clearPingInterval);
+  }
+}
+
+module.exports = { ioredisClient, keyvRedisClient, GLOBAL_PREFIX_SEPARATOR };

api/config/index.js

@@ -1,27 +1,13 @@
+const { MCPManager, FlowStateManager } = require('@librechat/api');
 const { EventSource } = require('eventsource');
 const { Time } = require('librechat-data-provider');
-const { MCPManager, FlowStateManager } = require('@librechat/api');
 const logger = require('./winston');
 
 global.EventSource = EventSource;
 
 /** @type {MCPManager} */
-let mcpManager = null;
 let flowManager = null;
 
-/**
- * @param {string} [userId] - Optional user ID, to avoid disconnecting the current user.
- * @returns {MCPManager}
- */
-function getMCPManager(userId) {
-  if (!mcpManager) {
-    mcpManager = MCPManager.getInstance(logger);
-  } else {
-    mcpManager.checkIdleConnections(userId);
-  }
-  return mcpManager;
-}
-
 /**
  * @param {Keyv} flowsCache
  * @returns {FlowStateManager}
@@ -30,7 +16,6 @@ function getFlowStateManager(flowsCache) {
   if (!flowManager) {
     flowManager = new FlowStateManager(flowsCache, {
       ttl: Time.ONE_MINUTE * 3,
-      logger,
     });
   }
   return flowManager;
@@ -38,6 +23,7 @@ function getFlowStateManager(flowsCache) {
 
 module.exports = {
   logger,
-  getMCPManager,
+  createMCPManager: MCPManager.createInstance,
+  getMCPManager: MCPManager.getInstance,
   getFlowStateManager,
 };

api/db/connect.js

@@ -1,11 +1,34 @@
 require('dotenv').config();
+const { isEnabled } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
+
 const mongoose = require('mongoose');
 const MONGO_URI = process.env.MONGO_URI;
 
 if (!MONGO_URI) {
   throw new Error('Please define the MONGO_URI environment variable');
 }
+/** The maximum number of connections in the connection pool. */
+const maxPoolSize = parseInt(process.env.MONGO_MAX_POOL_SIZE) || undefined;
+/** The minimum number of connections in the connection pool. */
+const minPoolSize = parseInt(process.env.MONGO_MIN_POOL_SIZE) || undefined;
+/** The maximum number of connections that may be in the process of being established concurrently by the connection pool. */
+const maxConnecting = parseInt(process.env.MONGO_MAX_CONNECTING) || undefined;
+/** The maximum number of milliseconds that a connection can remain idle in the pool before being removed and closed. */
+const maxIdleTimeMS = parseInt(process.env.MONGO_MAX_IDLE_TIME_MS) || undefined;
+/** The maximum time in milliseconds that a thread can wait for a connection to become available. */
+const waitQueueTimeoutMS = parseInt(process.env.MONGO_WAIT_QUEUE_TIMEOUT_MS) || undefined;
+/** Set to false to disable automatic index creation for all models associated with this connection. */
+const autoIndex =
+  process.env.MONGO_AUTO_INDEX != undefined
+    ? isEnabled(process.env.MONGO_AUTO_INDEX) || false
+    : undefined;
 
+/** Set to `false` to disable Mongoose automatically calling `createCollection()` on every model created on this connection. */
+const autoCreate =
+  process.env.MONGO_AUTO_CREATE != undefined
+    ? isEnabled(process.env.MONGO_AUTO_CREATE) || false
+    : undefined;
 /**
  * Global is used here to maintain a cached connection across hot reloads
  * in development. This prevents connections growing exponentially
@@ -26,13 +49,21 @@ async function connectDb() {
   if (!cached.promise || disconnected) {
     const opts = {
       bufferCommands: false,
+      ...(maxPoolSize ? { maxPoolSize } : {}),
+      ...(minPoolSize ? { minPoolSize } : {}),
+      ...(maxConnecting ? { maxConnecting } : {}),
+      ...(maxIdleTimeMS ? { maxIdleTimeMS } : {}),
+      ...(waitQueueTimeoutMS ? { waitQueueTimeoutMS } : {}),
+      ...(autoIndex != undefined ? { autoIndex } : {}),
+      ...(autoCreate != undefined ? { autoCreate } : {}),
       // useNewUrlParser: true,
       // useUnifiedTopology: true,
       // bufferMaxEntries: 0,
       // useFindAndModify: true,
       // useCreateIndex: true
     };
-
+    logger.info('Mongo Connection options');
+    logger.info(JSON.stringify(opts, null, 2));
     mongoose.set('strictQuery', true);
     cached.promise = mongoose.connect(MONGO_URI, opts).then((mongoose) => {
       return mongoose;

api/db/indexSync.js

@@ -1,8 +1,11 @@
 const mongoose = require('mongoose');
 const { MeiliSearch } = require('meilisearch');
 const { logger } = require('@librechat/data-schemas');
+const { FlowStateManager } = require('@librechat/api');
+const { CacheKeys } = require('librechat-data-provider');
 
 const { isEnabled } = require('~/server/utils');
+const { getLogStores } = require('~/cache');
 
 const Conversation = mongoose.models.Conversation;
 const Message = mongoose.models.Message;
@@ -28,43 +31,123 @@ class MeiliSearchClient {
   }
 }
 
+/**
+ * Performs the actual sync operations for messages and conversations
+ */
+async function performSync() {
+  const client = MeiliSearchClient.getInstance();
+
+  const { status } = await client.health();
+  if (status !== 'available') {
+    throw new Error('Meilisearch not available');
+  }
+
+  if (indexingDisabled === true) {
+    logger.info('[indexSync] Indexing is disabled, skipping...');
+    return { messagesSync: false, convosSync: false };
+  }
+
+  let messagesSync = false;
+  let convosSync = false;
+
+  // Check if we need to sync messages
+  const messageProgress = await Message.getSyncProgress();
+  if (!messageProgress.isComplete) {
+    logger.info(
+      `[indexSync] Messages need syncing: ${messageProgress.totalProcessed}/${messageProgress.totalDocuments} indexed`,
+    );
+
+    // Check if we should do a full sync or incremental
+    const messageCount = await Message.countDocuments();
+    const messagesIndexed = messageProgress.totalProcessed;
+    const syncThreshold = parseInt(process.env.MEILI_SYNC_THRESHOLD || '1000', 10);
+
+    if (messageCount - messagesIndexed > syncThreshold) {
+      logger.info('[indexSync] Starting full message sync due to large difference');
+      await Message.syncWithMeili();
+      messagesSync = true;
+    } else if (messageCount !== messagesIndexed) {
+      logger.warn('[indexSync] Messages out of sync, performing incremental sync');
+      await Message.syncWithMeili();
+      messagesSync = true;
+    }
+  } else {
+    logger.info(
+      `[indexSync] Messages are fully synced: ${messageProgress.totalProcessed}/${messageProgress.totalDocuments}`,
+    );
+  }
+
+  // Check if we need to sync conversations
+  const convoProgress = await Conversation.getSyncProgress();
+  if (!convoProgress.isComplete) {
+    logger.info(
+      `[indexSync] Conversations need syncing: ${convoProgress.totalProcessed}/${convoProgress.totalDocuments} indexed`,
+    );
+
+    const convoCount = await Conversation.countDocuments();
+    const convosIndexed = convoProgress.totalProcessed;
+    const syncThreshold = parseInt(process.env.MEILI_SYNC_THRESHOLD || '1000', 10);
+
+    if (convoCount - convosIndexed > syncThreshold) {
+      logger.info('[indexSync] Starting full conversation sync due to large difference');
+      await Conversation.syncWithMeili();
+      convosSync = true;
+    } else if (convoCount !== convosIndexed) {
+      logger.warn('[indexSync] Convos out of sync, performing incremental sync');
+      await Conversation.syncWithMeili();
+      convosSync = true;
+    }
+  } else {
+    logger.info(
+      `[indexSync] Conversations are fully synced: ${convoProgress.totalProcessed}/${convoProgress.totalDocuments}`,
+    );
+  }
+
+  return { messagesSync, convosSync };
+}
+
+/**
+ * Main index sync function that uses FlowStateManager to prevent concurrent execution
+ */
 async function indexSync() {
   if (!searchEnabled) {
     return;
   }
-  try {
-    const client = MeiliSearchClient.getInstance();
 
-    const { status } = await client.health();
-    if (status !== 'available') {
-      throw new Error('Meilisearch not available');
+  logger.info('[indexSync] Starting index synchronization check...');
+
+  try {
+    // Get or create FlowStateManager instance
+    const flowsCache = getLogStores(CacheKeys.FLOWS);
+    if (!flowsCache) {
+      logger.warn('[indexSync] Flows cache not available, falling back to direct sync');
+      return await performSync();
     }
 
-    if (indexingDisabled === true) {
-      logger.info('[indexSync] Indexing is disabled, skipping...');
+    const flowManager = new FlowStateManager(flowsCache, {
+      ttl: 60000 * 10, // 10 minutes TTL for sync operations
+    });
+
+    // Use a unique flow ID for the sync operation
+    const flowId = 'meili-index-sync';
+    const flowType = 'MEILI_SYNC';
+
+    // This will only execute the handler if no other instance is running the sync
+    const result = await flowManager.createFlowWithHandler(flowId, flowType, performSync);
+
+    if (result.messagesSync || result.convosSync) {
+      logger.info('[indexSync] Sync completed successfully');
+    } else {
+      logger.debug('[indexSync] No sync was needed');
+    }
+
+    return result;
+  } catch (err) {
+    if (err.message.includes('flow already exists')) {
+      logger.info('[indexSync] Sync already running on another instance');
       return;
     }
 
-    const messageCount = await Message.countDocuments();
-    const convoCount = await Conversation.countDocuments();
-    const messages = await client.index('messages').getStats();
-    const convos = await client.index('convos').getStats();
-    const messagesIndexed = messages.numberOfDocuments;
-    const convosIndexed = convos.numberOfDocuments;
-
-    logger.debug(`[indexSync] There are ${messageCount} messages and ${messagesIndexed} indexed`);
-    logger.debug(`[indexSync] There are ${convoCount} convos and ${convosIndexed} indexed`);
-
-    if (messageCount !== messagesIndexed) {
-      logger.debug('[indexSync] Messages out of sync, indexing');
-      Message.syncWithMeili();
-    }
-
-    if (convoCount !== convosIndexed) {
-      logger.debug('[indexSync] Convos out of sync, indexing');
-      Conversation.syncWithMeili();
-    }
-  } catch (err) {
     if (err.message.includes('not found')) {
       logger.debug('[indexSync] Creating indices...');
       currentTimeout = setTimeout(async () => {

api/jest.config.js

@@ -3,6 +3,7 @@ module.exports = {
   clearMocks: true,
   roots: ['<rootDir>'],
   coverageDirectory: 'coverage',
+  testTimeout: 30000, // 30 seconds timeout for all tests
   setupFiles: [
     './test/jestSetup.js',
     './test/__mocks__/logger.js',

api/models/Agent.js

@@ -1,17 +1,17 @@
 const mongoose = require('mongoose');
 const crypto = require('node:crypto');
 const { logger } = require('@librechat/data-schemas');
-const { SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
-const { GLOBAL_PROJECT_NAME, EPHEMERAL_AGENT_ID, mcp_delimiter } =
+const { ResourceType, SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
+const { GLOBAL_PROJECT_NAME, EPHEMERAL_AGENT_ID, mcp_all, mcp_delimiter } =
   require('librechat-data-provider').Constants;
-const { CONFIG_STORE, STARTUP_CONFIG } = require('librechat-data-provider').CacheKeys;
 const {
-  getProjectByName,
-  addAgentIdsToProject,
-  removeAgentIdsFromProject,
   removeAgentFromAllProjects,
+  removeAgentIdsFromProject,
+  addAgentIdsToProject,
+  getProjectByName,
 } = require('./Project');
-const getLogStores = require('~/cache/getLogStores');
+const { removeAllPermissions } = require('~/server/services/PermissionService');
+const { getCachedTools } = require('~/server/services/Config');
 const { getActions } = require('./Action');
 const { Agent } = require('~/db/models');
 
@@ -22,7 +22,7 @@ const { Agent } = require('~/db/models');
  * @throws {Error} If the agent creation fails.
  */
 const createAgent = async (agentData) => {
-  const { author, ...versionData } = agentData;
+  const { author: _author, ...versionData } = agentData;
   const timestamp = new Date();
   const initialAgentData = {
     ...agentData,
@@ -33,7 +33,9 @@ const createAgent = async (agentData) => {
         updatedAt: timestamp,
       },
     ],
+    category: agentData.category || 'general',
   };
+
   return (await Agent.create(initialAgentData)).toObject();
 };
 
@@ -55,12 +57,12 @@ const getAgent = async (searchParameter) => await Agent.findOne(searchParameter)
  * @param {string} params.agent_id
  * @param {string} params.endpoint
  * @param {import('@librechat/agents').ClientOptions} [params.model_parameters]
- * @returns {Agent|null} The agent document as a plain object, or null if not found.
+ * @returns {Promise<Agent|null>} The agent document as a plain object, or null if not found.
  */
-const loadEphemeralAgent = ({ req, agent_id, endpoint, model_parameters: _m }) => {
+const loadEphemeralAgent = async ({ req, agent_id, endpoint, model_parameters: _m }) => {
   const { model, ...model_parameters } = _m;
   /** @type {Record<string, FunctionTool>} */
-  const availableTools = req.app.locals.availableTools;
+  const availableTools = await getCachedTools({ userId: req.user.id, includeGlobal: true });
   /** @type {TEphemeralAgent | null} */
   const ephemeralAgent = req.body.ephemeralAgent;
   const mcpServers = new Set(ephemeralAgent?.mcp);
@@ -69,10 +71,14 @@ const loadEphemeralAgent = ({ req, agent_id, endpoint, model_parameters: _m }) =
   if (ephemeralAgent?.execute_code === true) {
     tools.push(Tools.execute_code);
   }
+  if (ephemeralAgent?.file_search === true) {
+    tools.push(Tools.file_search);
+  }
   if (ephemeralAgent?.web_search === true) {
     tools.push(Tools.web_search);
   }
 
+  const addedServers = new Set();
   if (mcpServers.size > 0) {
     for (const toolName of Object.keys(availableTools)) {
       if (!toolName.includes(mcp_delimiter)) {
@@ -80,13 +86,21 @@ const loadEphemeralAgent = ({ req, agent_id, endpoint, model_parameters: _m }) =
       }
       const mcpServer = toolName.split(mcp_delimiter)?.[1];
       if (mcpServer && mcpServers.has(mcpServer)) {
+        addedServers.add(mcpServer);
         tools.push(toolName);
       }
     }
+
+    for (const mcpServer of mcpServers) {
+      if (addedServers.has(mcpServer)) {
+        continue;
+      }
+      tools.push(`${mcp_all}${mcp_delimiter}${mcpServer}`);
+    }
   }
 
   const instructions = req.body.promptPrefix;
-  return {
+  const result = {
     id: agent_id,
     instructions,
     provider: endpoint,
@@ -94,6 +108,11 @@ const loadEphemeralAgent = ({ req, agent_id, endpoint, model_parameters: _m }) =
     model,
     tools,
   };
+
+  if (ephemeralAgent?.artifacts != null && ephemeralAgent.artifacts) {
+    result.artifacts = ephemeralAgent.artifacts;
+  }
+  return result;
 };
 
 /**
@@ -111,7 +130,7 @@ const loadAgent = async ({ req, agent_id, endpoint, model_parameters }) => {
     return null;
   }
   if (agent_id === EPHEMERAL_AGENT_ID) {
-    return loadEphemeralAgent({ req, agent_id, endpoint, model_parameters });
+    return await loadEphemeralAgent({ req, agent_id, endpoint, model_parameters });
   }
   const agent = await getAgent({
     id: agent_id,
@@ -122,29 +141,7 @@ const loadAgent = async ({ req, agent_id, endpoint, model_parameters }) => {
   }
 
   agent.version = agent.versions ? agent.versions.length : 0;
-
-  if (agent.author.toString() === req.user.id) {
-    return agent;
-  }
-
-  if (!agent.projectIds) {
-    return null;
-  }
-
-  const cache = getLogStores(CONFIG_STORE);
-  /** @type {TStartupConfig} */
-  const cachedStartupConfig = await cache.get(STARTUP_CONFIG);
-  let { instanceProjectId } = cachedStartupConfig ?? {};
-  if (!instanceProjectId) {
-    instanceProjectId = (await getProjectByName(GLOBAL_PROJECT_NAME, '_id'))._id.toString();
-  }
-
-  for (const projectObjectId of agent.projectIds) {
-    const projectId = projectObjectId.toString();
-    if (projectId === instanceProjectId) {
-      return agent;
-    }
-  }
+  return agent;
 };
 
 /**
@@ -174,7 +171,7 @@ const isDuplicateVersion = (updateData, currentData, versions, actionsHash = nul
     'actionsHash', // Exclude actionsHash from direct comparison
   ];
 
-  const { $push, $pull, $addToSet, ...directUpdates } = updateData;
+  const { $push: _$push, $pull: _$pull, $addToSet: _$addToSet, ...directUpdates } = updateData;
 
   if (Object.keys(directUpdates).length === 0 && !actionsHash) {
     return null;
@@ -193,54 +190,116 @@ const isDuplicateVersion = (updateData, currentData, versions, actionsHash = nul
 
   let isMatch = true;
   for (const field of importantFields) {
-    if (!wouldBeVersion[field] && !lastVersion[field]) {
+    const wouldBeValue = wouldBeVersion[field];
+    const lastVersionValue = lastVersion[field];
+
+    // Skip if both are undefined/null
+    if (!wouldBeValue && !lastVersionValue) {
       continue;
     }
 
-    if (Array.isArray(wouldBeVersion[field]) && Array.isArray(lastVersion[field])) {
-      if (wouldBeVersion[field].length !== lastVersion[field].length) {
+    // Handle arrays
+    if (Array.isArray(wouldBeValue) || Array.isArray(lastVersionValue)) {
+      // Normalize: treat undefined/null as empty array for comparison
+      let wouldBeArr;
+      if (Array.isArray(wouldBeValue)) {
+        wouldBeArr = wouldBeValue;
+      } else if (wouldBeValue == null) {
+        wouldBeArr = [];
+      } else {
+        wouldBeArr = [wouldBeValue];
+      }
+
+      let lastVersionArr;
+      if (Array.isArray(lastVersionValue)) {
+        lastVersionArr = lastVersionValue;
+      } else if (lastVersionValue == null) {
+        lastVersionArr = [];
+      } else {
+        lastVersionArr = [lastVersionValue];
+      }
+
+      if (wouldBeArr.length !== lastVersionArr.length) {
         isMatch = false;
         break;
       }
 
       // Special handling for projectIds (MongoDB ObjectIds)
       if (field === 'projectIds') {
-        const wouldBeIds = wouldBeVersion[field].map((id) => id.toString()).sort();
-        const versionIds = lastVersion[field].map((id) => id.toString()).sort();
+        const wouldBeIds = wouldBeArr.map((id) => id.toString()).sort();
+        const versionIds = lastVersionArr.map((id) => id.toString()).sort();
 
         if (!wouldBeIds.every((id, i) => id === versionIds[i])) {
           isMatch = false;
           break;
         }
       }
-      // Handle arrays of objects like tool_kwargs
-      else if (typeof wouldBeVersion[field][0] === 'object' && wouldBeVersion[field][0] !== null) {
-        const sortedWouldBe = [...wouldBeVersion[field]].map((item) => JSON.stringify(item)).sort();
-        const sortedVersion = [...lastVersion[field]].map((item) => JSON.stringify(item)).sort();
+      // Handle arrays of objects
+      else if (
+        wouldBeArr.length > 0 &&
+        typeof wouldBeArr[0] === 'object' &&
+        wouldBeArr[0] !== null
+      ) {
+        const sortedWouldBe = [...wouldBeArr].map((item) => JSON.stringify(item)).sort();
+        const sortedVersion = [...lastVersionArr].map((item) => JSON.stringify(item)).sort();
 
         if (!sortedWouldBe.every((item, i) => item === sortedVersion[i])) {
           isMatch = false;
           break;
         }
       } else {
-        const sortedWouldBe = [...wouldBeVersion[field]].sort();
-        const sortedVersion = [...lastVersion[field]].sort();
+        const sortedWouldBe = [...wouldBeArr].sort();
+        const sortedVersion = [...lastVersionArr].sort();
 
         if (!sortedWouldBe.every((item, i) => item === sortedVersion[i])) {
           isMatch = false;
           break;
         }
       }
-    } else if (field === 'model_parameters') {
-      const wouldBeParams = wouldBeVersion[field] || {};
-      const lastVersionParams = lastVersion[field] || {};
-      if (JSON.stringify(wouldBeParams) !== JSON.stringify(lastVersionParams)) {
+    }
+    // Handle objects
+    else if (typeof wouldBeValue === 'object' && wouldBeValue !== null) {
+      const lastVersionObj =
+        typeof lastVersionValue === 'object' && lastVersionValue !== null ? lastVersionValue : {};
+
+      // For empty objects, normalize the comparison
+      const wouldBeKeys = Object.keys(wouldBeValue);
+      const lastVersionKeys = Object.keys(lastVersionObj);
+
+      // If both are empty objects, they're equal
+      if (wouldBeKeys.length === 0 && lastVersionKeys.length === 0) {
+        continue;
+      }
+
+      // Otherwise do a deep comparison
+      if (JSON.stringify(wouldBeValue) !== JSON.stringify(lastVersionObj)) {
+        isMatch = false;
+        break;
+      }
+    }
+    // Handle primitive values
+    else {
+      // For primitives, handle the case where one is undefined and the other is a default value
+      if (wouldBeValue !== lastVersionValue) {
+        // Special handling for boolean false vs undefined
+        if (
+          typeof wouldBeValue === 'boolean' &&
+          wouldBeValue === false &&
+          lastVersionValue === undefined
+        ) {
+          continue;
+        }
+        // Special handling for empty string vs undefined
+        if (
+          typeof wouldBeValue === 'string' &&
+          wouldBeValue === '' &&
+          lastVersionValue === undefined
+        ) {
+          continue;
+        }
         isMatch = false;
         break;
       }
-    } else if (wouldBeVersion[field] !== lastVersion[field]) {
-      isMatch = false;
-      break;
     }
   }
 
@@ -269,7 +328,14 @@ const updateAgent = async (searchParameter, updateData, options = {}) => {
 
   const currentAgent = await Agent.findOne(searchParameter);
   if (currentAgent) {
-    const { __v, _id, id, versions, author, ...versionData } = currentAgent.toObject();
+    const {
+      __v,
+      _id,
+      id: __id,
+      versions,
+      author: _author,
+      ...versionData
+    } = currentAgent.toObject();
     const { $push, $pull, $addToSet, ...directUpdates } = updateData;
 
     let actionsHash = null;
@@ -307,17 +373,10 @@ const updateAgent = async (searchParameter, updateData, options = {}) => {
     if (shouldCreateVersion) {
       const duplicateVersion = isDuplicateVersion(updateData, versionData, versions, actionsHash);
       if (duplicateVersion && !forceVersion) {
-        const error = new Error(
-          'Duplicate version: This would create a version identical to an existing one',
-        );
-        error.statusCode = 409;
-        error.details = {
-          duplicateVersion,
-          versionIndex: versions.findIndex(
-            (v) => JSON.stringify(duplicateVersion) === JSON.stringify(v),
-          ),
-        };
-        throw error;
+        // No changes detected, return the current agent without creating a new version
+        const agentObj = currentAgent.toObject();
+        agentObj.version = versions.length;
+        return agentObj;
       }
     }
 
@@ -456,12 +515,117 @@ const deleteAgent = async (searchParameter) => {
   const agent = await Agent.findOneAndDelete(searchParameter);
   if (agent) {
     await removeAgentFromAllProjects(agent.id);
+    await removeAllPermissions({
+      resourceType: ResourceType.AGENT,
+      resourceId: agent._id,
+    });
   }
   return agent;
 };
 
+/**
+ * Get agents by accessible IDs with optional cursor-based pagination.
+ * @param {Object} params - The parameters for getting accessible agents.
+ * @param {Array} [params.accessibleIds] - Array of agent ObjectIds the user has ACL access to.
+ * @param {Object} [params.otherParams] - Additional query parameters (including author filter).
+ * @param {number} [params.limit] - Number of agents to return (max 100). If not provided, returns all agents.
+ * @param {string} [params.after] - Cursor for pagination - get agents after this cursor. // base64 encoded JSON string with updatedAt and _id.
+ * @returns {Promise<Object>} A promise that resolves to an object containing the agents data and pagination info.
+ */
+const getListAgentsByAccess = async ({
+  accessibleIds = [],
+  otherParams = {},
+  limit = null,
+  after = null,
+}) => {
+  const isPaginated = limit !== null && limit !== undefined;
+  const normalizedLimit = isPaginated ? Math.min(Math.max(1, parseInt(limit) || 20), 100) : null;
+
+  // Build base query combining ACL accessible agents with other filters
+  const baseQuery = { ...otherParams, _id: { $in: accessibleIds } };
+
+  // Add cursor condition
+  if (after) {
+    try {
+      const cursor = JSON.parse(Buffer.from(after, 'base64').toString('utf8'));
+      const { updatedAt, _id } = cursor;
+
+      const cursorCondition = {
+        $or: [
+          { updatedAt: { $lt: new Date(updatedAt) } },
+          { updatedAt: new Date(updatedAt), _id: { $gt: new mongoose.Types.ObjectId(_id) } },
+        ],
+      };
+
+      // Merge cursor condition with base query
+      if (Object.keys(baseQuery).length > 0) {
+        baseQuery.$and = [{ ...baseQuery }, cursorCondition];
+        // Remove the original conditions from baseQuery to avoid duplication
+        Object.keys(baseQuery).forEach((key) => {
+          if (key !== '$and') delete baseQuery[key];
+        });
+      } else {
+        Object.assign(baseQuery, cursorCondition);
+      }
+    } catch (error) {
+      logger.warn('Invalid cursor:', error.message);
+    }
+  }
+
+  let query = Agent.find(baseQuery, {
+    id: 1,
+    _id: 1,
+    name: 1,
+    avatar: 1,
+    author: 1,
+    projectIds: 1,
+    description: 1,
+    updatedAt: 1,
+    category: 1,
+    support_contact: 1,
+    is_promoted: 1,
+  }).sort({ updatedAt: -1, _id: 1 });
+
+  // Only apply limit if pagination is requested
+  if (isPaginated) {
+    query = query.limit(normalizedLimit + 1);
+  }
+
+  const agents = await query.lean();
+
+  const hasMore = isPaginated ? agents.length > normalizedLimit : false;
+  const data = (isPaginated ? agents.slice(0, normalizedLimit) : agents).map((agent) => {
+    if (agent.author) {
+      agent.author = agent.author.toString();
+    }
+    return agent;
+  });
+
+  // Generate next cursor only if paginated
+  let nextCursor = null;
+  if (isPaginated && hasMore && data.length > 0) {
+    const lastAgent = agents[normalizedLimit - 1];
+    nextCursor = Buffer.from(
+      JSON.stringify({
+        updatedAt: lastAgent.updatedAt.toISOString(),
+        _id: lastAgent._id.toString(),
+      }),
+    ).toString('base64');
+  }
+
+  return {
+    object: 'list',
+    data,
+    first_id: data.length > 0 ? data[0].id : null,
+    last_id: data.length > 0 ? data[data.length - 1].id : null,
+    has_more: hasMore,
+    after: nextCursor,
+  };
+};
+
 /**
  * Get all agents.
+ * @deprecated Use getListAgentsByAccess for ACL-aware agent listing
  * @param {Object} searchParameter - The search parameters to find matching agents.
  * @param {string} searchParameter.author - The user ID of the agent's author.
  * @returns {Promise<Object>} A promise that resolves to an object containing the agents data and pagination info.
@@ -480,13 +644,15 @@ const getListAgents = async (searchParameter) => {
   const agents = (
     await Agent.find(query, {
       id: 1,
-      _id: 0,
+      _id: 1,
       name: 1,
       avatar: 1,
       author: 1,
       projectIds: 1,
       description: 1,
+      // @deprecated - isCollaborative replaced by ACL permissions
       isCollaborative: 1,
+      category: 1,
     }).lean()
   ).map((agent) => {
     if (agent.author?.toString() !== author) {
@@ -515,7 +681,7 @@ const getListAgents = async (searchParameter) => {
  * This function also updates the corresponding projects to include or exclude the agent ID.
  *
  * @param {Object} params - Parameters for updating the agent's projects.
- * @param {MongoUser} params.user - Parameters for updating the agent's projects.
+ * @param {IUser} params.user - Parameters for updating the agent's projects.
  * @param {string} params.agentId - The ID of the agent to update.
  * @param {string[]} [params.projectIds] - Array of project IDs to add to the agent.
  * @param {string[]} [params.removeProjectIds] - Array of project IDs to remove from the agent.
@@ -652,6 +818,14 @@ const generateActionMetadataHash = async (actionIds, actions) => {
 
   return hashHex;
 };
+/**
+ * Counts the number of promoted agents.
+ * @returns  {Promise<number>} - The count of promoted agents
+ */
+const countPromotedAgents = async () => {
+  const count = await Agent.countDocuments({ is_promoted: true });
+  return count;
+};
 
 /**
  * Load a default agent based on the endpoint
@@ -669,6 +843,8 @@ module.exports = {
   revertAgentVersion,
   updateAgentProjects,
   addAgentResourceFile,
+  getListAgentsByAccess,
   removeAgentResourceFiles,
   generateActionMetadataHash,
+  countPromotedAgents,
 };

api/models/Conversation.js

@@ -1,4 +1,5 @@
 const { logger } = require('@librechat/data-schemas');
+const { createTempChatExpirationDate } = require('@librechat/api');
 const { getMessages, deleteMessages } = require('./Message');
 const { Conversation } = require('~/db/models');
 
@@ -98,10 +99,15 @@ module.exports = {
         update.conversationId = newConversationId;
       }
 
-      if (req.body.isTemporary) {
-        const expiredAt = new Date();
-        expiredAt.setDate(expiredAt.getDate() + 30);
-        update.expiredAt = expiredAt;
+      if (req?.body?.isTemporary) {
+        try {
+          const appConfig = req.config;
+          update.expiredAt = createTempChatExpirationDate(appConfig?.interfaceConfig);
+        } catch (err) {
+          logger.error('Error creating temporary chat expiration date:', err);
+          logger.info(`---\`saveConvo\` context: ${metadata?.context}`);
+          update.expiredAt = null;
+        }
       } else {
         update.expiredAt = null;
       }

api/models/Conversation.spec.js

@@ -0,0 +1,570 @@
+const mongoose = require('mongoose');
+const { v4: uuidv4 } = require('uuid');
+const { EModelEndpoint } = require('librechat-data-provider');
+const { MongoMemoryServer } = require('mongodb-memory-server');
+const {
+  deleteNullOrEmptyConversations,
+  searchConversation,
+  getConvosByCursor,
+  getConvosQueried,
+  getConvoFiles,
+  getConvoTitle,
+  deleteConvos,
+  saveConvo,
+  getConvo,
+} = require('./Conversation');
+jest.mock('~/server/services/Config/app');
+jest.mock('./Message');
+const { getMessages, deleteMessages } = require('./Message');
+
+const { Conversation } = require('~/db/models');
+
+describe('Conversation Operations', () => {
+  let mongoServer;
+  let mockReq;
+  let mockConversationData;
+
+  beforeAll(async () => {
+    mongoServer = await MongoMemoryServer.create();
+    const mongoUri = mongoServer.getUri();
+    await mongoose.connect(mongoUri);
+  });
+
+  afterAll(async () => {
+    await mongoose.disconnect();
+    await mongoServer.stop();
+  });
+
+  beforeEach(async () => {
+    // Clear database
+    await Conversation.deleteMany({});
+
+    // Reset mocks
+    jest.clearAllMocks();
+
+    // Default mock implementations
+    getMessages.mockResolvedValue([]);
+    deleteMessages.mockResolvedValue({ deletedCount: 0 });
+
+    mockReq = {
+      user: { id: 'user123' },
+      body: {},
+      config: {
+        interfaceConfig: {
+          temporaryChatRetention: 24, // Default 24 hours
+        },
+      },
+    };
+
+    mockConversationData = {
+      conversationId: uuidv4(),
+      title: 'Test Conversation',
+      endpoint: EModelEndpoint.openAI,
+    };
+  });
+
+  describe('saveConvo', () => {
+    it('should save a conversation for an authenticated user', async () => {
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.user).toBe('user123');
+      expect(result.title).toBe('Test Conversation');
+      expect(result.endpoint).toBe(EModelEndpoint.openAI);
+
+      // Verify the conversation was actually saved to the database
+      const savedConvo = await Conversation.findOne({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+      });
+      expect(savedConvo).toBeTruthy();
+      expect(savedConvo.title).toBe('Test Conversation');
+    });
+
+    it('should query messages when saving a conversation', async () => {
+      // Mock messages as ObjectIds
+      const mongoose = require('mongoose');
+      const mockMessages = [new mongoose.Types.ObjectId(), new mongoose.Types.ObjectId()];
+      getMessages.mockResolvedValue(mockMessages);
+
+      await saveConvo(mockReq, mockConversationData);
+
+      // Verify that getMessages was called with correct parameters
+      expect(getMessages).toHaveBeenCalledWith(
+        { conversationId: mockConversationData.conversationId },
+        '_id',
+      );
+    });
+
+    it('should handle newConversationId when provided', async () => {
+      const newConversationId = uuidv4();
+      const result = await saveConvo(mockReq, {
+        ...mockConversationData,
+        newConversationId,
+      });
+
+      expect(result.conversationId).toBe(newConversationId);
+    });
+
+    it('should handle unsetFields metadata', async () => {
+      const metadata = {
+        unsetFields: { someField: 1 },
+      };
+
+      await saveConvo(mockReq, mockConversationData, metadata);
+
+      const savedConvo = await Conversation.findOne({
+        conversationId: mockConversationData.conversationId,
+      });
+      expect(savedConvo.someField).toBeUndefined();
+    });
+  });
+
+  describe('isTemporary conversation handling', () => {
+    it('should save a conversation with expiredAt when isTemporary is true', async () => {
+      // Mock app config with 24 hour retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 24;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+      const afterSave = new Date();
+
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.expiredAt).toBeDefined();
+      expect(result.expiredAt).toBeInstanceOf(Date);
+
+      // Verify expiredAt is approximately 24 hours in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 24 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        new Date(afterSave.getTime() + 24 * 60 * 60 * 1000 + 1000).getTime(),
+      );
+    });
+
+    it('should save a conversation without expiredAt when isTemporary is false', async () => {
+      mockReq.body = { isTemporary: false };
+
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should save a conversation without expiredAt when isTemporary is not provided', async () => {
+      // No isTemporary in body
+      mockReq.body = {};
+
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should use custom retention period from config', async () => {
+      // Mock app config with 48 hour retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 48;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 48 hours in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 48 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle minimum retention period (1 hour)', async () => {
+      // Mock app config with less than minimum retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 0.5; // Half hour - should be clamped to 1 hour
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 1 hour in the future (minimum)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 1 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle maximum retention period (8760 hours)', async () => {
+      // Mock app config with more than maximum retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 10000; // Should be clamped to 8760 hours
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 8760 hours (1 year) in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 8760 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle missing config gracefully', async () => {
+      // Simulate missing config - should use default retention period
+      delete mockReq.config;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+      const afterSave = new Date();
+
+      // Should still save the conversation with default retention period (30 days)
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.expiredAt).toBeDefined();
+      expect(result.expiredAt).toBeInstanceOf(Date);
+
+      // Verify expiredAt is approximately 30 days in the future (720 hours)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 720 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        new Date(afterSave.getTime() + 720 * 60 * 60 * 1000 + 1000).getTime(),
+      );
+    });
+
+    it('should use default retention when config is not provided', async () => {
+      // Mock getAppConfig to return empty config
+      mockReq.config = {}; // Empty config
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Default retention is 30 days (720 hours)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 30 * 24 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should update expiredAt when saving existing temporary conversation', async () => {
+      // First save a temporary conversation
+      mockReq.config.interfaceConfig.temporaryChatRetention = 24;
+
+      mockReq.body = { isTemporary: true };
+      const firstSave = await saveConvo(mockReq, mockConversationData);
+      const originalExpiredAt = firstSave.expiredAt;
+
+      // Wait a bit to ensure time difference
+      await new Promise((resolve) => setTimeout(resolve, 100));
+
+      // Save again with same conversationId but different title
+      const updatedData = { ...mockConversationData, title: 'Updated Title' };
+      const secondSave = await saveConvo(mockReq, updatedData);
+
+      // Should update title and create new expiredAt
+      expect(secondSave.title).toBe('Updated Title');
+      expect(secondSave.expiredAt).toBeDefined();
+      expect(new Date(secondSave.expiredAt).getTime()).toBeGreaterThan(
+        new Date(originalExpiredAt).getTime(),
+      );
+    });
+
+    it('should not set expiredAt when updating non-temporary conversation', async () => {
+      // First save a non-temporary conversation
+      mockReq.body = { isTemporary: false };
+      const firstSave = await saveConvo(mockReq, mockConversationData);
+      expect(firstSave.expiredAt).toBeNull();
+
+      // Update without isTemporary flag
+      mockReq.body = {};
+      const updatedData = { ...mockConversationData, title: 'Updated Title' };
+      const secondSave = await saveConvo(mockReq, updatedData);
+
+      expect(secondSave.title).toBe('Updated Title');
+      expect(secondSave.expiredAt).toBeNull();
+    });
+
+    it('should filter out expired conversations in getConvosByCursor', async () => {
+      // Create some test conversations
+      const nonExpiredConvo = await Conversation.create({
+        conversationId: uuidv4(),
+        user: 'user123',
+        title: 'Non-expired',
+        endpoint: EModelEndpoint.openAI,
+        expiredAt: null,
+        updatedAt: new Date(),
+      });
+
+      await Conversation.create({
+        conversationId: uuidv4(),
+        user: 'user123',
+        title: 'Future expired',
+        endpoint: EModelEndpoint.openAI,
+        expiredAt: new Date(Date.now() + 24 * 60 * 60 * 1000), // 24 hours from now
+        updatedAt: new Date(),
+      });
+
+      // Mock Meili search
+      Conversation.meiliSearch = jest.fn().mockResolvedValue({ hits: [] });
+
+      const result = await getConvosByCursor('user123');
+
+      // Should only return conversations with null or non-existent expiredAt
+      expect(result.conversations).toHaveLength(1);
+      expect(result.conversations[0].conversationId).toBe(nonExpiredConvo.conversationId);
+    });
+
+    it('should filter out expired conversations in getConvosQueried', async () => {
+      // Create test conversations
+      const nonExpiredConvo = await Conversation.create({
+        conversationId: uuidv4(),
+        user: 'user123',
+        title: 'Non-expired',
+        endpoint: EModelEndpoint.openAI,
+        expiredAt: null,
+      });
+
+      const expiredConvo = await Conversation.create({
+        conversationId: uuidv4(),
+        user: 'user123',
+        title: 'Expired',
+        endpoint: EModelEndpoint.openAI,
+        expiredAt: new Date(Date.now() + 24 * 60 * 60 * 1000),
+      });
+
+      const convoIds = [
+        { conversationId: nonExpiredConvo.conversationId },
+        { conversationId: expiredConvo.conversationId },
+      ];
+
+      const result = await getConvosQueried('user123', convoIds);
+
+      // Should only return the non-expired conversation
+      expect(result.conversations).toHaveLength(1);
+      expect(result.conversations[0].conversationId).toBe(nonExpiredConvo.conversationId);
+      expect(result.convoMap[nonExpiredConvo.conversationId]).toBeDefined();
+      expect(result.convoMap[expiredConvo.conversationId]).toBeUndefined();
+    });
+  });
+
+  describe('searchConversation', () => {
+    it('should find a conversation by conversationId', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        title: 'Test',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      const result = await searchConversation(mockConversationData.conversationId);
+
+      expect(result).toBeTruthy();
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.user).toBe('user123');
+      expect(result.title).toBeUndefined(); // Only returns conversationId and user
+    });
+
+    it('should return null if conversation not found', async () => {
+      const result = await searchConversation('non-existent-id');
+      expect(result).toBeNull();
+    });
+  });
+
+  describe('getConvo', () => {
+    it('should retrieve a conversation for a user', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        title: 'Test Conversation',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      const result = await getConvo('user123', mockConversationData.conversationId);
+
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.user).toBe('user123');
+      expect(result.title).toBe('Test Conversation');
+    });
+
+    it('should return null if conversation not found', async () => {
+      const result = await getConvo('user123', 'non-existent-id');
+      expect(result).toBeNull();
+    });
+  });
+
+  describe('getConvoTitle', () => {
+    it('should return the conversation title', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        title: 'Test Title',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      const result = await getConvoTitle('user123', mockConversationData.conversationId);
+      expect(result).toBe('Test Title');
+    });
+
+    it('should return null if conversation has no title', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        title: null,
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      const result = await getConvoTitle('user123', mockConversationData.conversationId);
+      expect(result).toBeNull();
+    });
+
+    it('should return "New Chat" if conversation not found', async () => {
+      const result = await getConvoTitle('user123', 'non-existent-id');
+      expect(result).toBe('New Chat');
+    });
+  });
+
+  describe('getConvoFiles', () => {
+    it('should return conversation files', async () => {
+      const files = ['file1', 'file2'];
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        endpoint: EModelEndpoint.openAI,
+        files,
+      });
+
+      const result = await getConvoFiles(mockConversationData.conversationId);
+      expect(result).toEqual(files);
+    });
+
+    it('should return empty array if no files', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      const result = await getConvoFiles(mockConversationData.conversationId);
+      expect(result).toEqual([]);
+    });
+
+    it('should return empty array if conversation not found', async () => {
+      const result = await getConvoFiles('non-existent-id');
+      expect(result).toEqual([]);
+    });
+  });
+
+  describe('deleteConvos', () => {
+    it('should delete conversations and associated messages', async () => {
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user123',
+        title: 'To Delete',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      deleteMessages.mockResolvedValue({ deletedCount: 5 });
+
+      const result = await deleteConvos('user123', {
+        conversationId: mockConversationData.conversationId,
+      });
+
+      expect(result.deletedCount).toBe(1);
+      expect(result.messages.deletedCount).toBe(5);
+      expect(deleteMessages).toHaveBeenCalledWith({
+        conversationId: { $in: [mockConversationData.conversationId] },
+      });
+
+      // Verify conversation was deleted
+      const deletedConvo = await Conversation.findOne({
+        conversationId: mockConversationData.conversationId,
+      });
+      expect(deletedConvo).toBeNull();
+    });
+
+    it('should throw error if no conversations found', async () => {
+      await expect(deleteConvos('user123', { conversationId: 'non-existent' })).rejects.toThrow(
+        'Conversation not found or already deleted.',
+      );
+    });
+  });
+
+  describe('deleteNullOrEmptyConversations', () => {
+    it('should delete conversations with null, empty, or missing conversationIds', async () => {
+      // Since conversationId is required by the schema, we can't create documents with null/missing IDs
+      // This test should verify the function works when such documents exist (e.g., from data corruption)
+
+      // For this test, let's create a valid conversation and verify the function doesn't delete it
+      await Conversation.create({
+        conversationId: mockConversationData.conversationId,
+        user: 'user4',
+        endpoint: EModelEndpoint.openAI,
+      });
+
+      deleteMessages.mockResolvedValue({ deletedCount: 0 });
+
+      const result = await deleteNullOrEmptyConversations();
+
+      expect(result.conversations.deletedCount).toBe(0); // No invalid conversations to delete
+      expect(result.messages.deletedCount).toBe(0);
+
+      // Verify valid conversation remains
+      const remainingConvos = await Conversation.find({});
+      expect(remainingConvos).toHaveLength(1);
+      expect(remainingConvos[0].conversationId).toBe(mockConversationData.conversationId);
+    });
+  });
+
+  describe('Error Handling', () => {
+    it('should handle database errors in saveConvo', async () => {
+      // Force a database error by disconnecting
+      await mongoose.disconnect();
+
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      expect(result).toEqual({ message: 'Error saving conversation' });
+
+      // Reconnect for other tests
+      await mongoose.connect(mongoServer.getUri());
+    });
+  });
+});

api/models/File.js

@@ -1,5 +1,5 @@
 const { logger } = require('@librechat/data-schemas');
-const { EToolResources } = require('librechat-data-provider');
+const { EToolResources, FileContext } = require('librechat-data-provider');
 const { File } = require('~/db/models');
 
 /**
@@ -32,19 +32,19 @@ const getFiles = async (filter, _sortOptions, selectFields = { text: 0 }) => {
  * @returns {Promise<Array<MongoFile>>} Files that match the criteria
  */
 const getToolFilesByIds = async (fileIds, toolResourceSet) => {
-  if (!fileIds || !fileIds.length) {
+  if (!fileIds || !fileIds.length || !toolResourceSet?.size) {
     return [];
   }
 
   try {
     const filter = {
       file_id: { $in: fileIds },
+      $or: [],
     };
 
-    if (toolResourceSet.size) {
-      filter.$or = [];
+    if (toolResourceSet.has(EToolResources.ocr)) {
+      filter.$or.push({ text: { $exists: true, $ne: null }, context: FileContext.agents });
     }
-
     if (toolResourceSet.has(EToolResources.file_search)) {
       filter.$or.push({ embedded: true });
     }

api/models/File.spec.js

@@ -0,0 +1,621 @@
+const mongoose = require('mongoose');
+const { v4: uuidv4 } = require('uuid');
+const { createModels } = require('@librechat/data-schemas');
+const { MongoMemoryServer } = require('mongodb-memory-server');
+const {
+  SystemRoles,
+  ResourceType,
+  AccessRoleIds,
+  PrincipalType,
+} = require('librechat-data-provider');
+const { grantPermission } = require('~/server/services/PermissionService');
+const { getFiles, createFile } = require('./File');
+const { seedDefaultRoles } = require('~/models');
+const { createAgent } = require('./Agent');
+
+let File;
+let Agent;
+let AclEntry;
+let User;
+let modelsToCleanup = [];
+
+describe('File Access Control', () => {
+  let mongoServer;
+
+  beforeAll(async () => {
+    mongoServer = await MongoMemoryServer.create();
+    const mongoUri = mongoServer.getUri();
+    await mongoose.connect(mongoUri);
+
+    // Initialize all models
+    const models = createModels(mongoose);
+
+    // Track which models we're adding
+    modelsToCleanup = Object.keys(models);
+
+    // Register models on mongoose.models so methods can access them
+    const dbModels = require('~/db/models');
+    Object.assign(mongoose.models, dbModels);
+
+    File = dbModels.File;
+    Agent = dbModels.Agent;
+    AclEntry = dbModels.AclEntry;
+    User = dbModels.User;
+
+    // Seed default roles
+    await seedDefaultRoles();
+  });
+
+  afterAll(async () => {
+    // Clean up all collections before disconnecting
+    const collections = mongoose.connection.collections;
+    for (const key in collections) {
+      await collections[key].deleteMany({});
+    }
+
+    // Clear only the models we added
+    for (const modelName of modelsToCleanup) {
+      if (mongoose.models[modelName]) {
+        delete mongoose.models[modelName];
+      }
+    }
+
+    await mongoose.disconnect();
+    await mongoServer.stop();
+  });
+
+  beforeEach(async () => {
+    await File.deleteMany({});
+    await Agent.deleteMany({});
+    await AclEntry.deleteMany({});
+    await User.deleteMany({});
+    // Don't delete AccessRole as they are seeded defaults needed for tests
+  });
+
+  describe('hasAccessToFilesViaAgent', () => {
+    it('should efficiently check access for multiple files at once', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileIds = [uuidv4(), uuidv4(), uuidv4(), uuidv4()];
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create files
+      for (const fileId of fileIds) {
+        await createFile({
+          user: authorId,
+          file_id: fileId,
+          filename: `file-${fileId}.txt`,
+          filepath: `/uploads/${fileId}`,
+        });
+      }
+
+      // Create agent with only first two files attached
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Test Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: [fileIds[0], fileIds[1]],
+          },
+        },
+      });
+
+      // Grant EDIT permission to user on the agent
+      await grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: userId,
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_EDITOR,
+        grantedBy: authorId,
+      });
+
+      // Check access for all files
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMap = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: SystemRoles.USER,
+        fileIds,
+        agentId: agent.id, // Use agent.id which is the custom UUID
+      });
+
+      // Should have access only to the first two files
+      expect(accessMap.get(fileIds[0])).toBe(true);
+      expect(accessMap.get(fileIds[1])).toBe(true);
+      expect(accessMap.get(fileIds[2])).toBe(false);
+      expect(accessMap.get(fileIds[3])).toBe(false);
+    });
+
+    it('should grant access to all files when user is the agent author', async () => {
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileIds = [uuidv4(), uuidv4(), uuidv4()];
+
+      // Create author user
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create agent
+      await createAgent({
+        id: agentId,
+        name: 'Test Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: [fileIds[0]], // Only one file attached
+          },
+        },
+      });
+
+      // Check access as the author
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMap = await hasAccessToFilesViaAgent({
+        userId: authorId,
+        role: SystemRoles.USER,
+        fileIds,
+        agentId,
+      });
+
+      // Author should have access to all files
+      expect(accessMap.get(fileIds[0])).toBe(true);
+      expect(accessMap.get(fileIds[1])).toBe(true);
+      expect(accessMap.get(fileIds[2])).toBe(true);
+    });
+
+    it('should handle non-existent agent gracefully', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const fileIds = [uuidv4(), uuidv4()];
+
+      // Create user
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMap = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: SystemRoles.USER,
+        fileIds,
+        agentId: 'non-existent-agent',
+      });
+
+      // Should have no access to any files
+      expect(accessMap.get(fileIds[0])).toBe(false);
+      expect(accessMap.get(fileIds[1])).toBe(false);
+    });
+
+    it('should deny access when user only has VIEW permission and needs access for deletion', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileIds = [uuidv4(), uuidv4()];
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create agent with files
+      const agent = await createAgent({
+        id: agentId,
+        name: 'View-Only Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: fileIds,
+          },
+        },
+      });
+
+      // Grant only VIEW permission to user on the agent
+      await grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: userId,
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_VIEWER,
+        grantedBy: authorId,
+      });
+
+      // Check access for files
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMap = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: SystemRoles.USER,
+        fileIds,
+        agentId,
+        isDelete: true,
+      });
+
+      // Should have no access to any files when only VIEW permission
+      expect(accessMap.get(fileIds[0])).toBe(false);
+      expect(accessMap.get(fileIds[1])).toBe(false);
+    });
+
+    it('should grant access when user has VIEW permission', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileIds = [uuidv4(), uuidv4()];
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create agent with files
+      const agent = await createAgent({
+        id: agentId,
+        name: 'View-Only Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: fileIds,
+          },
+        },
+      });
+
+      // Grant only VIEW permission to user on the agent
+      await grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: userId,
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_VIEWER,
+        grantedBy: authorId,
+      });
+
+      // Check access for files
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMap = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: SystemRoles.USER,
+        fileIds,
+        agentId,
+      });
+
+      expect(accessMap.get(fileIds[0])).toBe(true);
+      expect(accessMap.get(fileIds[1])).toBe(true);
+    });
+  });
+
+  describe('getFiles with agent access control', () => {
+    test('should return files owned by user and files accessible through agent', async () => {
+      const authorId = new mongoose.Types.ObjectId();
+      const userId = new mongoose.Types.ObjectId();
+      const agentId = `agent_${uuidv4()}`;
+      const ownedFileId = `file_${uuidv4()}`;
+      const sharedFileId = `file_${uuidv4()}`;
+      const inaccessibleFileId = `file_${uuidv4()}`;
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create agent with shared file
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Shared Agent',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+        tool_resources: {
+          file_search: {
+            file_ids: [sharedFileId],
+          },
+        },
+      });
+
+      // Grant EDIT permission to user on the agent
+      await grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: userId,
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_EDITOR,
+        grantedBy: authorId,
+      });
+
+      // Create files
+      await createFile({
+        file_id: ownedFileId,
+        user: userId,
+        filename: 'owned.txt',
+        filepath: '/uploads/owned.txt',
+        type: 'text/plain',
+        bytes: 100,
+      });
+
+      await createFile({
+        file_id: sharedFileId,
+        user: authorId,
+        filename: 'shared.txt',
+        filepath: '/uploads/shared.txt',
+        type: 'text/plain',
+        bytes: 200,
+        embedded: true,
+      });
+
+      await createFile({
+        file_id: inaccessibleFileId,
+        user: authorId,
+        filename: 'inaccessible.txt',
+        filepath: '/uploads/inaccessible.txt',
+        type: 'text/plain',
+        bytes: 300,
+      });
+
+      // Get all files first
+      const allFiles = await getFiles(
+        { file_id: { $in: [ownedFileId, sharedFileId, inaccessibleFileId] } },
+        null,
+        { text: 0 },
+      );
+
+      // Then filter by access control
+      const { filterFilesByAgentAccess } = require('~/server/services/Files/permissions');
+      const files = await filterFilesByAgentAccess({
+        files: allFiles,
+        userId: userId,
+        role: SystemRoles.USER,
+        agentId,
+      });
+
+      expect(files).toHaveLength(2);
+      expect(files.map((f) => f.file_id)).toContain(ownedFileId);
+      expect(files.map((f) => f.file_id)).toContain(sharedFileId);
+      expect(files.map((f) => f.file_id)).not.toContain(inaccessibleFileId);
+    });
+
+    test('should return all files when no userId/agentId provided', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const fileId1 = `file_${uuidv4()}`;
+      const fileId2 = `file_${uuidv4()}`;
+
+      await createFile({
+        file_id: fileId1,
+        user: userId,
+        filename: 'file1.txt',
+        filepath: '/uploads/file1.txt',
+        type: 'text/plain',
+        bytes: 100,
+      });
+
+      await createFile({
+        file_id: fileId2,
+        user: new mongoose.Types.ObjectId(),
+        filename: 'file2.txt',
+        filepath: '/uploads/file2.txt',
+        type: 'text/plain',
+        bytes: 200,
+      });
+
+      const files = await getFiles({ file_id: { $in: [fileId1, fileId2] } });
+      expect(files).toHaveLength(2);
+    });
+  });
+
+  describe('Role-based file permissions', () => {
+    it('should optimize permission checks when role is provided', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileIds = [uuidv4(), uuidv4()];
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+        role: 'ADMIN', // User has ADMIN role
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create files
+      for (const fileId of fileIds) {
+        await createFile({
+          file_id: fileId,
+          user: authorId,
+          filename: `${fileId}.txt`,
+          filepath: `/uploads/${fileId}.txt`,
+          type: 'text/plain',
+          bytes: 100,
+        });
+      }
+
+      // Create agent with files
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Test Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: fileIds,
+          },
+        },
+      });
+
+      // Grant permission to ADMIN role
+      await grantPermission({
+        principalType: PrincipalType.ROLE,
+        principalId: 'ADMIN',
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_EDITOR,
+        grantedBy: authorId,
+      });
+
+      // Check access with role provided (should avoid DB query)
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+      const accessMapWithRole = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: 'ADMIN',
+        fileIds,
+        agentId: agent.id,
+      });
+
+      // User should have access through their ADMIN role
+      expect(accessMapWithRole.get(fileIds[0])).toBe(true);
+      expect(accessMapWithRole.get(fileIds[1])).toBe(true);
+
+      // Check access without role (will query DB to get user's role)
+      const accessMapWithoutRole = await hasAccessToFilesViaAgent({
+        userId: userId,
+        fileIds,
+        agentId: agent.id,
+      });
+
+      // Should have same result
+      expect(accessMapWithoutRole.get(fileIds[0])).toBe(true);
+      expect(accessMapWithoutRole.get(fileIds[1])).toBe(true);
+    });
+
+    it('should deny access when user role changes', async () => {
+      const userId = new mongoose.Types.ObjectId();
+      const authorId = new mongoose.Types.ObjectId();
+      const agentId = uuidv4();
+      const fileId = uuidv4();
+
+      // Create users
+      await User.create({
+        _id: userId,
+        email: 'user@example.com',
+        emailVerified: true,
+        provider: 'local',
+        role: 'EDITOR',
+      });
+
+      await User.create({
+        _id: authorId,
+        email: 'author@example.com',
+        emailVerified: true,
+        provider: 'local',
+      });
+
+      // Create file
+      await createFile({
+        file_id: fileId,
+        user: authorId,
+        filename: 'test.txt',
+        filepath: '/uploads/test.txt',
+        type: 'text/plain',
+        bytes: 100,
+      });
+
+      // Create agent
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Test Agent',
+        author: authorId,
+        model: 'gpt-4',
+        provider: 'openai',
+        tool_resources: {
+          file_search: {
+            file_ids: [fileId],
+          },
+        },
+      });
+
+      // Grant permission to EDITOR role only
+      await grantPermission({
+        principalType: PrincipalType.ROLE,
+        principalId: 'EDITOR',
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_EDITOR,
+        grantedBy: authorId,
+      });
+
+      const { hasAccessToFilesViaAgent } = require('~/server/services/Files/permissions');
+
+      // Check with EDITOR role - should have access
+      const accessAsEditor = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: 'EDITOR',
+        fileIds: [fileId],
+        agentId: agent.id,
+      });
+      expect(accessAsEditor.get(fileId)).toBe(true);
+
+      // Simulate role change to USER - should lose access
+      const accessAsUser = await hasAccessToFilesViaAgent({
+        userId: userId,
+        role: SystemRoles.USER,
+        fileIds: [fileId],
+        agentId: agent.id,
+      });
+      expect(accessAsUser.get(fileId)).toBe(false);
+    });
+  });
+});

api/models/Message.js

@@ -1,5 +1,6 @@
 const { z } = require('zod');
 const { logger } = require('@librechat/data-schemas');
+const { createTempChatExpirationDate } = require('@librechat/api');
 const { Message } = require('~/db/models');
 
 const idSchema = z.string().uuid();
@@ -9,7 +10,7 @@ const idSchema = z.string().uuid();
  *
  * @async
  * @function saveMessage
- * @param {Express.Request} req - The request object containing user information.
+ * @param {ServerRequest} req - The request object containing user information.
  * @param {Object} params - The message data object.
  * @param {string} params.endpoint - The endpoint where the message originated.
  * @param {string} params.iconURL - The URL of the sender's icon.
@@ -54,9 +55,14 @@ async function saveMessage(req, params, metadata) {
     };
 
     if (req?.body?.isTemporary) {
-      const expiredAt = new Date();
-      expiredAt.setDate(expiredAt.getDate() + 30);
-      update.expiredAt = expiredAt;
+      try {
+        const appConfig = req.config;
+        update.expiredAt = createTempChatExpirationDate(appConfig?.interfaceConfig);
+      } catch (err) {
+        logger.error('Error creating temporary chat expiration date:', err);
+        logger.info(`---\`saveMessage\` context: ${metadata?.context}`);
+        update.expiredAt = null;
+      }
     } else {
       update.expiredAt = null;
     }

api/models/Message.spec.js

@@ -1,17 +1,20 @@
 const mongoose = require('mongoose');
-const { MongoMemoryServer } = require('mongodb-memory-server');
 const { v4: uuidv4 } = require('uuid');
 const { messageSchema } = require('@librechat/data-schemas');
+const { MongoMemoryServer } = require('mongodb-memory-server');
 
 const {
   saveMessage,
   getMessages,
   updateMessage,
   deleteMessages,
+  bulkSaveMessages,
   updateMessageText,
   deleteMessagesSince,
 } = require('./Message');
 
+jest.mock('~/server/services/Config/app');
+
 /**
  * @type {import('mongoose').Model<import('@librechat/data-schemas').IMessage>}
  */
@@ -40,6 +43,11 @@ describe('Message Operations', () => {
 
     mockReq = {
       user: { id: 'user123' },
+      config: {
+        interfaceConfig: {
+          temporaryChatRetention: 24, // Default 24 hours
+        },
+      },
     };
 
     mockMessageData = {
@@ -117,21 +125,21 @@ describe('Message Operations', () => {
       const conversationId = uuidv4();
 
       // Create multiple messages in the same conversation
-      const message1 = await saveMessage(mockReq, {
+      await saveMessage(mockReq, {
         messageId: 'msg1',
         conversationId,
         text: 'First message',
         user: 'user123',
       });
 
-      const message2 = await saveMessage(mockReq, {
+      await saveMessage(mockReq, {
         messageId: 'msg2',
         conversationId,
         text: 'Second message',
         user: 'user123',
       });
 
-      const message3 = await saveMessage(mockReq, {
+      await saveMessage(mockReq, {
         messageId: 'msg3',
         conversationId,
         text: 'Third message',
@@ -314,4 +322,255 @@ describe('Message Operations', () => {
       expect(messages[0].text).toBe('Victim message');
     });
   });
+
+  describe('isTemporary message handling', () => {
+    beforeEach(() => {
+      // Reset mocks before each test
+      jest.clearAllMocks();
+    });
+
+    it('should save a message with expiredAt when isTemporary is true', async () => {
+      // Mock app config with 24 hour retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 24;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+      const afterSave = new Date();
+
+      expect(result.messageId).toBe('msg123');
+      expect(result.expiredAt).toBeDefined();
+      expect(result.expiredAt).toBeInstanceOf(Date);
+
+      // Verify expiredAt is approximately 24 hours in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 24 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        new Date(afterSave.getTime() + 24 * 60 * 60 * 1000 + 1000).getTime(),
+      );
+    });
+
+    it('should save a message without expiredAt when isTemporary is false', async () => {
+      mockReq.body = { isTemporary: false };
+
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.messageId).toBe('msg123');
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should save a message without expiredAt when isTemporary is not provided', async () => {
+      // No isTemporary in body
+      mockReq.body = {};
+
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.messageId).toBe('msg123');
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should use custom retention period from config', async () => {
+      // Mock app config with 48 hour retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 48;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 48 hours in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 48 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle minimum retention period (1 hour)', async () => {
+      // Mock app config with less than minimum retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 0.5; // Half hour - should be clamped to 1 hour
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 1 hour in the future (minimum)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 1 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle maximum retention period (8760 hours)', async () => {
+      // Mock app config with more than maximum retention
+      mockReq.config.interfaceConfig.temporaryChatRetention = 10000; // Should be clamped to 8760 hours
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Verify expiredAt is approximately 8760 hours (1 year) in the future
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 8760 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should handle missing config gracefully', async () => {
+      // Simulate missing config - should use default retention period
+      delete mockReq.config;
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+      const afterSave = new Date();
+
+      // Should still save the message with default retention period (30 days)
+      expect(result.messageId).toBe('msg123');
+      expect(result.expiredAt).toBeDefined();
+      expect(result.expiredAt).toBeInstanceOf(Date);
+
+      // Verify expiredAt is approximately 30 days in the future (720 hours)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 720 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        new Date(afterSave.getTime() + 720 * 60 * 60 * 1000 + 1000).getTime(),
+      );
+    });
+
+    it('should use default retention when config is not provided', async () => {
+      // Mock getAppConfig to return empty config
+      mockReq.config = {}; // Empty config
+
+      mockReq.body = { isTemporary: true };
+
+      const beforeSave = new Date();
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      expect(result.expiredAt).toBeDefined();
+
+      // Default retention is 30 days (720 hours)
+      const expectedExpirationTime = new Date(beforeSave.getTime() + 30 * 24 * 60 * 60 * 1000);
+      const actualExpirationTime = new Date(result.expiredAt);
+
+      expect(actualExpirationTime.getTime()).toBeGreaterThanOrEqual(
+        expectedExpirationTime.getTime() - 1000,
+      );
+      expect(actualExpirationTime.getTime()).toBeLessThanOrEqual(
+        expectedExpirationTime.getTime() + 1000,
+      );
+    });
+
+    it('should not update expiredAt on message update', async () => {
+      // First save a temporary message
+      mockReq.config.interfaceConfig.temporaryChatRetention = 24;
+
+      mockReq.body = { isTemporary: true };
+      const savedMessage = await saveMessage(mockReq, mockMessageData);
+      const originalExpiredAt = savedMessage.expiredAt;
+
+      // Now update the message without isTemporary flag
+      mockReq.body = {};
+      const updatedMessage = await updateMessage(mockReq, {
+        messageId: 'msg123',
+        text: 'Updated text',
+      });
+
+      // expiredAt should not be in the returned updated message object
+      expect(updatedMessage.expiredAt).toBeUndefined();
+
+      // Verify in database that expiredAt wasn't changed
+      const dbMessage = await Message.findOne({ messageId: 'msg123', user: 'user123' });
+      expect(dbMessage.expiredAt).toEqual(originalExpiredAt);
+    });
+
+    it('should preserve expiredAt when saving existing temporary message', async () => {
+      // First save a temporary message
+      mockReq.config.interfaceConfig.temporaryChatRetention = 24;
+
+      mockReq.body = { isTemporary: true };
+      const firstSave = await saveMessage(mockReq, mockMessageData);
+      const originalExpiredAt = firstSave.expiredAt;
+
+      // Wait a bit to ensure time difference
+      await new Promise((resolve) => setTimeout(resolve, 100));
+
+      // Save again with same messageId but different text
+      const updatedData = { ...mockMessageData, text: 'Updated text' };
+      const secondSave = await saveMessage(mockReq, updatedData);
+
+      // Should update text but create new expiredAt
+      expect(secondSave.text).toBe('Updated text');
+      expect(secondSave.expiredAt).toBeDefined();
+      expect(new Date(secondSave.expiredAt).getTime()).toBeGreaterThan(
+        new Date(originalExpiredAt).getTime(),
+      );
+    });
+
+    it('should handle bulk operations with temporary messages', async () => {
+      // This test verifies bulkSaveMessages doesn't interfere with expiredAt
+      const messages = [
+        {
+          messageId: 'bulk1',
+          conversationId: uuidv4(),
+          text: 'Bulk message 1',
+          user: 'user123',
+          expiredAt: new Date(Date.now() + 24 * 60 * 60 * 1000),
+        },
+        {
+          messageId: 'bulk2',
+          conversationId: uuidv4(),
+          text: 'Bulk message 2',
+          user: 'user123',
+          expiredAt: null,
+        },
+      ];
+
+      await bulkSaveMessages(messages);
+
+      const savedMessages = await Message.find({
+        messageId: { $in: ['bulk1', 'bulk2'] },
+      }).lean();
+
+      expect(savedMessages).toHaveLength(2);
+
+      const bulk1 = savedMessages.find((m) => m.messageId === 'bulk1');
+      const bulk2 = savedMessages.find((m) => m.messageId === 'bulk2');
+
+      expect(bulk1.expiredAt).toBeDefined();
+      expect(bulk2.expiredAt).toBeNull();
+    });
+  });
 });

api/models/Prompt.js

@@ -1,12 +1,18 @@
 const { ObjectId } = require('mongodb');
 const { logger } = require('@librechat/data-schemas');
-const { SystemRoles, SystemCategories, Constants } = require('librechat-data-provider');
 const {
-  getProjectByName,
-  addGroupIdsToProject,
-  removeGroupIdsFromProject,
+  Constants,
+  SystemRoles,
+  ResourceType,
+  SystemCategories,
+} = require('librechat-data-provider');
+const {
   removeGroupFromAllProjects,
+  removeGroupIdsFromProject,
+  addGroupIdsToProject,
+  getProjectByName,
 } = require('./Project');
+const { removeAllPermissions } = require('~/server/services/PermissionService');
 const { PromptGroup, Prompt } = require('~/db/models');
 const { escapeRegExp } = require('~/server/utils');
 
@@ -100,10 +106,6 @@ const getAllPromptGroups = async (req, filter) => {
   try {
     const { name, ...query } = filter;
 
-    if (!query.author) {
-      throw new Error('Author is required');
-    }
-
     let searchShared = true;
     let searchSharedOnly = false;
     if (name) {
@@ -153,10 +155,6 @@ const getPromptGroups = async (req, filter) => {
     const validatedPageNumber = Math.max(parseInt(pageNumber, 10), 1);
     const validatedPageSize = Math.max(parseInt(pageSize, 10), 1);
 
-    if (!query.author) {
-      throw new Error('Author is required');
-    }
-
     let searchShared = true;
     let searchSharedOnly = false;
     if (name) {
@@ -221,12 +219,16 @@ const getPromptGroups = async (req, filter) => {
  * @returns {Promise<TDeletePromptGroupResponse>}
  */
 const deletePromptGroup = async ({ _id, author, role }) => {
-  const query = { _id, author };
-  const groupQuery = { groupId: new ObjectId(_id), author };
-  if (role === SystemRoles.ADMIN) {
-    delete query.author;
-    delete groupQuery.author;
+  // Build query - with ACL, author is optional
+  const query = { _id };
+  const groupQuery = { groupId: new ObjectId(_id) };
+
+  // Legacy: Add author filter if provided (backward compatibility)
+  if (author && role !== SystemRoles.ADMIN) {
+    query.author = author;
+    groupQuery.author = author;
   }
+
   const response = await PromptGroup.deleteOne(query);
 
   if (!response || response.deletedCount === 0) {
@@ -235,13 +237,140 @@ const deletePromptGroup = async ({ _id, author, role }) => {
 
   await Prompt.deleteMany(groupQuery);
   await removeGroupFromAllProjects(_id);
+
+  try {
+    await removeAllPermissions({ resourceType: ResourceType.PROMPTGROUP, resourceId: _id });
+  } catch (error) {
+    logger.error('Error removing promptGroup permissions:', error);
+  }
+
   return { message: 'Prompt group deleted successfully' };
 };
 
+/**
+ * Get prompt groups by accessible IDs with optional cursor-based pagination.
+ * @param {Object} params - The parameters for getting accessible prompt groups.
+ * @param {Array} [params.accessibleIds] - Array of prompt group ObjectIds the user has ACL access to.
+ * @param {Object} [params.otherParams] - Additional query parameters (including author filter).
+ * @param {number} [params.limit] - Number of prompt groups to return (max 100). If not provided, returns all prompt groups.
+ * @param {string} [params.after] - Cursor for pagination - get prompt groups after this cursor. // base64 encoded JSON string with updatedAt and _id.
+ * @returns {Promise<Object>} A promise that resolves to an object containing the prompt groups data and pagination info.
+ */
+async function getListPromptGroupsByAccess({
+  accessibleIds = [],
+  otherParams = {},
+  limit = null,
+  after = null,
+}) {
+  const isPaginated = limit !== null && limit !== undefined;
+  const normalizedLimit = isPaginated ? Math.min(Math.max(1, parseInt(limit) || 20), 100) : null;
+
+  // Build base query combining ACL accessible prompt groups with other filters
+  const baseQuery = { ...otherParams, _id: { $in: accessibleIds } };
+
+  // Add cursor condition
+  if (after && typeof after === 'string' && after !== 'undefined' && after !== 'null') {
+    try {
+      const cursor = JSON.parse(Buffer.from(after, 'base64').toString('utf8'));
+      const { updatedAt, _id } = cursor;
+
+      const cursorCondition = {
+        $or: [
+          { updatedAt: { $lt: new Date(updatedAt) } },
+          { updatedAt: new Date(updatedAt), _id: { $gt: new ObjectId(_id) } },
+        ],
+      };
+
+      // Merge cursor condition with base query
+      if (Object.keys(baseQuery).length > 0) {
+        baseQuery.$and = [{ ...baseQuery }, cursorCondition];
+        // Remove the original conditions from baseQuery to avoid duplication
+        Object.keys(baseQuery).forEach((key) => {
+          if (key !== '$and') delete baseQuery[key];
+        });
+      } else {
+        Object.assign(baseQuery, cursorCondition);
+      }
+    } catch (error) {
+      logger.warn('Invalid cursor:', error.message);
+    }
+  }
+
+  // Build aggregation pipeline
+  const pipeline = [{ $match: baseQuery }, { $sort: { updatedAt: -1, _id: 1 } }];
+
+  // Only apply limit if pagination is requested
+  if (isPaginated) {
+    pipeline.push({ $limit: normalizedLimit + 1 });
+  }
+
+  // Add lookup for production prompt
+  pipeline.push(
+    {
+      $lookup: {
+        from: 'prompts',
+        localField: 'productionId',
+        foreignField: '_id',
+        as: 'productionPrompt',
+      },
+    },
+    { $unwind: { path: '$productionPrompt', preserveNullAndEmptyArrays: true } },
+    {
+      $project: {
+        name: 1,
+        numberOfGenerations: 1,
+        oneliner: 1,
+        category: 1,
+        projectIds: 1,
+        productionId: 1,
+        author: 1,
+        authorName: 1,
+        createdAt: 1,
+        updatedAt: 1,
+        'productionPrompt.prompt': 1,
+      },
+    },
+  );
+
+  const promptGroups = await PromptGroup.aggregate(pipeline).exec();
+
+  const hasMore = isPaginated ? promptGroups.length > normalizedLimit : false;
+  const data = (isPaginated ? promptGroups.slice(0, normalizedLimit) : promptGroups).map(
+    (group) => {
+      if (group.author) {
+        group.author = group.author.toString();
+      }
+      return group;
+    },
+  );
+
+  // Generate next cursor only if paginated
+  let nextCursor = null;
+  if (isPaginated && hasMore && data.length > 0) {
+    const lastGroup = promptGroups[normalizedLimit - 1];
+    nextCursor = Buffer.from(
+      JSON.stringify({
+        updatedAt: lastGroup.updatedAt.toISOString(),
+        _id: lastGroup._id.toString(),
+      }),
+    ).toString('base64');
+  }
+
+  return {
+    object: 'list',
+    data,
+    first_id: data.length > 0 ? data[0]._id.toString() : null,
+    last_id: data.length > 0 ? data[data.length - 1]._id.toString() : null,
+    has_more: hasMore,
+    after: nextCursor,
+  };
+}
+
 module.exports = {
   getPromptGroups,
   deletePromptGroup,
   getAllPromptGroups,
+  getListPromptGroupsByAccess,
   /**
    * Create a prompt and its respective group
    * @param {TCreatePromptRecord} saveData
@@ -430,6 +559,16 @@ module.exports = {
       .lean();
 
     if (remainingPrompts.length === 0) {
+      // Remove all ACL entries for the promptGroup when deleting the last prompt
+      try {
+        await removeAllPermissions({
+          resourceType: ResourceType.PROMPTGROUP,
+          resourceId: groupId,
+        });
+      } catch (error) {
+        logger.error('Error removing promptGroup permissions:', error);
+      }
+
       await PromptGroup.deleteOne({ _id: groupId });
       await removeGroupFromAllProjects(groupId);
 

api/models/Prompt.spec.js

@@ -0,0 +1,564 @@
+const mongoose = require('mongoose');
+const { ObjectId } = require('mongodb');
+const { logger } = require('@librechat/data-schemas');
+const { MongoMemoryServer } = require('mongodb-memory-server');
+const {
+  SystemRoles,
+  ResourceType,
+  AccessRoleIds,
+  PrincipalType,
+  PermissionBits,
+} = require('librechat-data-provider');
+
+// Mock the config/connect module to prevent connection attempts during tests
+jest.mock('../../config/connect', () => jest.fn().mockResolvedValue(true));
+
+const dbModels = require('~/db/models');
+
+// Disable console for tests
+logger.silent = true;
+
+let mongoServer;
+let Prompt, PromptGroup, AclEntry, AccessRole, User, Group, Project;
+let promptFns, permissionService;
+let testUsers, testGroups, testRoles;
+
+beforeAll(async () => {
+  // Set up MongoDB memory server
+  mongoServer = await MongoMemoryServer.create();
+  const mongoUri = mongoServer.getUri();
+  await mongoose.connect(mongoUri);
+
+  // Initialize models
+  Prompt = dbModels.Prompt;
+  PromptGroup = dbModels.PromptGroup;
+  AclEntry = dbModels.AclEntry;
+  AccessRole = dbModels.AccessRole;
+  User = dbModels.User;
+  Group = dbModels.Group;
+  Project = dbModels.Project;
+
+  promptFns = require('~/models/Prompt');
+  permissionService = require('~/server/services/PermissionService');
+
+  // Create test data
+  await setupTestData();
+});
+
+afterAll(async () => {
+  await mongoose.disconnect();
+  await mongoServer.stop();
+  jest.clearAllMocks();
+});
+
+async function setupTestData() {
+  // Create access roles for promptGroups
+  testRoles = {
+    viewer: await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_VIEWER,
+      name: 'Viewer',
+      description: 'Can view promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits: PermissionBits.VIEW,
+    }),
+    editor: await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_EDITOR,
+      name: 'Editor',
+      description: 'Can view and edit promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits: PermissionBits.VIEW | PermissionBits.EDIT,
+    }),
+    owner: await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+      name: 'Owner',
+      description: 'Full control over promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits:
+        PermissionBits.VIEW | PermissionBits.EDIT | PermissionBits.DELETE | PermissionBits.SHARE,
+    }),
+  };
+
+  // Create test users
+  testUsers = {
+    owner: await User.create({
+      name: 'Prompt Owner',
+      email: 'owner@example.com',
+      role: SystemRoles.USER,
+    }),
+    editor: await User.create({
+      name: 'Prompt Editor',
+      email: 'editor@example.com',
+      role: SystemRoles.USER,
+    }),
+    viewer: await User.create({
+      name: 'Prompt Viewer',
+      email: 'viewer@example.com',
+      role: SystemRoles.USER,
+    }),
+    admin: await User.create({
+      name: 'Admin User',
+      email: 'admin@example.com',
+      role: SystemRoles.ADMIN,
+    }),
+    noAccess: await User.create({
+      name: 'No Access User',
+      email: 'noaccess@example.com',
+      role: SystemRoles.USER,
+    }),
+  };
+
+  // Create test groups
+  testGroups = {
+    editors: await Group.create({
+      name: 'Prompt Editors',
+      description: 'Group with editor access',
+    }),
+    viewers: await Group.create({
+      name: 'Prompt Viewers',
+      description: 'Group with viewer access',
+    }),
+  };
+
+  await Project.create({
+    name: 'Global',
+    description: 'Global project',
+    promptGroupIds: [],
+  });
+}
+
+describe('Prompt ACL Permissions', () => {
+  describe('Creating Prompts with Permissions', () => {
+    it('should grant owner permissions when creating a prompt', async () => {
+      // First create a group
+      const testGroup = await PromptGroup.create({
+        name: 'Test Group',
+        category: 'testing',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new mongoose.Types.ObjectId(),
+      });
+
+      const promptData = {
+        prompt: {
+          prompt: 'Test prompt content',
+          name: 'Test Prompt',
+          type: 'text',
+          groupId: testGroup._id,
+        },
+        author: testUsers.owner._id,
+      };
+
+      await promptFns.savePrompt(promptData);
+
+      // Manually grant permissions as would happen in the route
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+        grantedBy: testUsers.owner._id,
+      });
+
+      // Check ACL entry
+      const aclEntry = await AclEntry.findOne({
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testGroup._id,
+        principalType: PrincipalType.USER,
+        principalId: testUsers.owner._id,
+      });
+
+      expect(aclEntry).toBeTruthy();
+      expect(aclEntry.permBits).toBe(testRoles.owner.permBits);
+    });
+  });
+
+  describe('Accessing Prompts', () => {
+    let testPromptGroup;
+
+    beforeEach(async () => {
+      // Create a prompt group
+      testPromptGroup = await PromptGroup.create({
+        name: 'Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      // Create a prompt
+      await Prompt.create({
+        prompt: 'Test prompt for access control',
+        name: 'Access Test Prompt',
+        author: testUsers.owner._id,
+        groupId: testPromptGroup._id,
+        type: 'text',
+      });
+
+      // Grant owner permissions
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+        grantedBy: testUsers.owner._id,
+      });
+    });
+
+    afterEach(async () => {
+      await Prompt.deleteMany({});
+      await PromptGroup.deleteMany({});
+      await AclEntry.deleteMany({});
+    });
+
+    it('owner should have full access to their prompt', async () => {
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+      });
+
+      expect(hasAccess).toBe(true);
+
+      const canEdit = await permissionService.checkPermission({
+        userId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.EDIT,
+      });
+
+      expect(canEdit).toBe(true);
+    });
+
+    it('user with viewer role should only have view access', async () => {
+      // Grant viewer permissions
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: testUsers.viewer._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_VIEWER,
+        grantedBy: testUsers.owner._id,
+      });
+
+      const canView = await permissionService.checkPermission({
+        userId: testUsers.viewer._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+      });
+
+      const canEdit = await permissionService.checkPermission({
+        userId: testUsers.viewer._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.EDIT,
+      });
+
+      expect(canView).toBe(true);
+      expect(canEdit).toBe(false);
+    });
+
+    it('user without permissions should have no access', async () => {
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.noAccess._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+      });
+
+      expect(hasAccess).toBe(false);
+    });
+
+    it('admin should have access regardless of permissions', async () => {
+      // Admin users should work through normal permission system
+      // The middleware layer handles admin bypass, not the permission service
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.admin._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+      });
+
+      // Without explicit permissions, even admin won't have access at this layer
+      expect(hasAccess).toBe(false);
+
+      // The actual admin bypass happens in the middleware layer (`canAccessPromptViaGroup`/`canAccessPromptGroupResource`)
+      // which checks req.user.role === SystemRoles.ADMIN
+    });
+  });
+
+  describe('Group-based Access', () => {
+    let testPromptGroup;
+
+    beforeEach(async () => {
+      // Create a prompt group first
+      testPromptGroup = await PromptGroup.create({
+        name: 'Group Access Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      await Prompt.create({
+        prompt: 'Group access test prompt',
+        name: 'Group Test',
+        author: testUsers.owner._id,
+        groupId: testPromptGroup._id,
+        type: 'text',
+      });
+
+      // Add users to groups
+      await User.findByIdAndUpdate(testUsers.editor._id, {
+        $push: { groups: testGroups.editors._id },
+      });
+
+      await User.findByIdAndUpdate(testUsers.viewer._id, {
+        $push: { groups: testGroups.viewers._id },
+      });
+    });
+
+    afterEach(async () => {
+      await Prompt.deleteMany({});
+      await AclEntry.deleteMany({});
+      await User.updateMany({}, { $set: { groups: [] } });
+    });
+
+    it('group members should inherit group permissions', async () => {
+      // Create a prompt group
+      const testPromptGroup = await PromptGroup.create({
+        name: 'Group Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      const { addUserToGroup } = require('~/models');
+      await addUserToGroup(testUsers.editor._id, testGroups.editors._id);
+
+      const prompt = await promptFns.savePrompt({
+        author: testUsers.owner._id,
+        prompt: {
+          prompt: 'Group test prompt',
+          name: 'Group Test',
+          groupId: testPromptGroup._id,
+          type: 'text',
+        },
+      });
+
+      // Check if savePrompt returned an error
+      if (!prompt || !prompt.prompt) {
+        throw new Error(`Failed to save prompt: ${prompt?.message || 'Unknown error'}`);
+      }
+
+      // Grant edit permissions to the group
+      await permissionService.grantPermission({
+        principalType: PrincipalType.GROUP,
+        principalId: testGroups.editors._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_EDITOR,
+        grantedBy: testUsers.owner._id,
+      });
+
+      // Check if group member has access
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.editor._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.EDIT,
+      });
+
+      expect(hasAccess).toBe(true);
+
+      // Check that non-member doesn't have access
+      const nonMemberAccess = await permissionService.checkPermission({
+        userId: testUsers.viewer._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        requiredPermission: PermissionBits.EDIT,
+      });
+
+      expect(nonMemberAccess).toBe(false);
+    });
+  });
+
+  describe('Public Access', () => {
+    let publicPromptGroup, privatePromptGroup;
+
+    beforeEach(async () => {
+      // Create separate prompt groups for public and private access
+      publicPromptGroup = await PromptGroup.create({
+        name: 'Public Access Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      privatePromptGroup = await PromptGroup.create({
+        name: 'Private Access Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      // Create prompts in their respective groups
+      await Prompt.create({
+        prompt: 'Public prompt',
+        name: 'Public',
+        author: testUsers.owner._id,
+        groupId: publicPromptGroup._id,
+        type: 'text',
+      });
+
+      await Prompt.create({
+        prompt: 'Private prompt',
+        name: 'Private',
+        author: testUsers.owner._id,
+        groupId: privatePromptGroup._id,
+        type: 'text',
+      });
+
+      // Grant public view access to publicPromptGroup
+      await permissionService.grantPermission({
+        principalType: PrincipalType.PUBLIC,
+        principalId: null,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: publicPromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_VIEWER,
+        grantedBy: testUsers.owner._id,
+      });
+
+      // Grant only owner access to privatePromptGroup
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: privatePromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+        grantedBy: testUsers.owner._id,
+      });
+    });
+
+    afterEach(async () => {
+      await Prompt.deleteMany({});
+      await PromptGroup.deleteMany({});
+      await AclEntry.deleteMany({});
+    });
+
+    it('public prompt should be accessible to any user', async () => {
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.noAccess._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: publicPromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+        includePublic: true,
+      });
+
+      expect(hasAccess).toBe(true);
+    });
+
+    it('private prompt should not be accessible to unauthorized users', async () => {
+      const hasAccess = await permissionService.checkPermission({
+        userId: testUsers.noAccess._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: privatePromptGroup._id,
+        requiredPermission: PermissionBits.VIEW,
+        includePublic: true,
+      });
+
+      expect(hasAccess).toBe(false);
+    });
+  });
+
+  describe('Prompt Deletion', () => {
+    let testPromptGroup;
+
+    it('should remove ACL entries when prompt is deleted', async () => {
+      testPromptGroup = await PromptGroup.create({
+        name: 'Deletion Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      const prompt = await promptFns.savePrompt({
+        author: testUsers.owner._id,
+        prompt: {
+          prompt: 'To be deleted',
+          name: 'Delete Test',
+          groupId: testPromptGroup._id,
+          type: 'text',
+        },
+      });
+
+      // Check if savePrompt returned an error
+      if (!prompt || !prompt.prompt) {
+        throw new Error(`Failed to save prompt: ${prompt?.message || 'Unknown error'}`);
+      }
+
+      const testPromptId = prompt.prompt._id;
+      const promptGroupId = testPromptGroup._id;
+
+      // Grant permission
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: testUsers.owner._id,
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+        accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+        grantedBy: testUsers.owner._id,
+      });
+
+      // Verify ACL entry exists
+      const beforeDelete = await AclEntry.find({
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+      });
+      expect(beforeDelete).toHaveLength(1);
+
+      // Delete the prompt
+      await promptFns.deletePrompt({
+        promptId: testPromptId,
+        groupId: promptGroupId,
+        author: testUsers.owner._id,
+        role: SystemRoles.USER,
+      });
+
+      // Verify ACL entries are removed
+      const aclEntries = await AclEntry.find({
+        resourceType: ResourceType.PROMPTGROUP,
+        resourceId: testPromptGroup._id,
+      });
+
+      expect(aclEntries).toHaveLength(0);
+    });
+  });
+
+  describe('Backwards Compatibility', () => {
+    it('should handle prompts without ACL entries gracefully', async () => {
+      // Create a prompt group first
+      const promptGroup = await PromptGroup.create({
+        name: 'Legacy Test Group',
+        author: testUsers.owner._id,
+        authorName: testUsers.owner.name,
+        productionId: new ObjectId(),
+      });
+
+      // Create a prompt without ACL entries (legacy prompt)
+      const legacyPrompt = await Prompt.create({
+        prompt: 'Legacy prompt without ACL',
+        name: 'Legacy',
+        author: testUsers.owner._id,
+        groupId: promptGroup._id,
+        type: 'text',
+      });
+
+      // The system should handle this gracefully
+      const prompt = await promptFns.getPrompt({ _id: legacyPrompt._id });
+      expect(prompt).toBeTruthy();
+      expect(prompt._id.toString()).toBe(legacyPrompt._id.toString());
+    });
+  });
+});

api/models/PromptGroupMigration.spec.js

@@ -0,0 +1,280 @@
+const mongoose = require('mongoose');
+const { ObjectId } = require('mongodb');
+const { logger } = require('@librechat/data-schemas');
+const { MongoMemoryServer } = require('mongodb-memory-server');
+const {
+  Constants,
+  ResourceType,
+  AccessRoleIds,
+  PrincipalType,
+  PrincipalModel,
+  PermissionBits,
+} = require('librechat-data-provider');
+
+// Mock the config/connect module to prevent connection attempts during tests
+jest.mock('../../config/connect', () => jest.fn().mockResolvedValue(true));
+
+// Disable console for tests
+logger.silent = true;
+
+describe('PromptGroup Migration Script', () => {
+  let mongoServer;
+  let Prompt, PromptGroup, AclEntry, AccessRole, User, Project;
+  let migrateToPromptGroupPermissions;
+  let testOwner, testProject;
+  let ownerRole, viewerRole;
+
+  beforeAll(async () => {
+    // Set up MongoDB memory server
+    mongoServer = await MongoMemoryServer.create();
+    const mongoUri = mongoServer.getUri();
+    await mongoose.connect(mongoUri);
+
+    // Initialize models
+    const dbModels = require('~/db/models');
+    Prompt = dbModels.Prompt;
+    PromptGroup = dbModels.PromptGroup;
+    AclEntry = dbModels.AclEntry;
+    AccessRole = dbModels.AccessRole;
+    User = dbModels.User;
+    Project = dbModels.Project;
+
+    // Create test user
+    testOwner = await User.create({
+      name: 'Test Owner',
+      email: 'owner@test.com',
+      role: 'USER',
+    });
+
+    // Create test project with the proper name
+    const projectName = Constants.GLOBAL_PROJECT_NAME || 'instance';
+    testProject = await Project.create({
+      name: projectName,
+      description: 'Global project',
+      promptGroupIds: [],
+    });
+
+    // Create promptGroup access roles
+    ownerRole = await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_OWNER,
+      name: 'Owner',
+      description: 'Full control over promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits:
+        PermissionBits.VIEW | PermissionBits.EDIT | PermissionBits.DELETE | PermissionBits.SHARE,
+    });
+
+    viewerRole = await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_VIEWER,
+      name: 'Viewer',
+      description: 'Can view promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits: PermissionBits.VIEW,
+    });
+
+    await AccessRole.create({
+      accessRoleId: AccessRoleIds.PROMPTGROUP_EDITOR,
+      name: 'Editor',
+      description: 'Can view and edit promptGroups',
+      resourceType: ResourceType.PROMPTGROUP,
+      permBits: PermissionBits.VIEW | PermissionBits.EDIT,
+    });
+
+    // Import migration function
+    const migration = require('../../config/migrate-prompt-permissions');
+    migrateToPromptGroupPermissions = migration.migrateToPromptGroupPermissions;
+  });
+
+  afterAll(async () => {
+    await mongoose.disconnect();
+    await mongoServer.stop();
+  });
+
+  beforeEach(async () => {
+    // Clean up before each test
+    await Prompt.deleteMany({});
+    await PromptGroup.deleteMany({});
+    await AclEntry.deleteMany({});
+    // Reset the project's promptGroupIds array
+    testProject.promptGroupIds = [];
+    await testProject.save();
+  });
+
+  it('should categorize promptGroups correctly in dry run', async () => {
+    // Create global prompt group (in Global project)
+    const globalPromptGroup = await PromptGroup.create({
+      name: 'Global Group',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    // Create private prompt group (not in any project)
+    await PromptGroup.create({
+      name: 'Private Group',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    // Add global group to project's promptGroupIds array
+    testProject.promptGroupIds = [globalPromptGroup._id];
+    await testProject.save();
+
+    const result = await migrateToPromptGroupPermissions({ dryRun: true });
+
+    expect(result.dryRun).toBe(true);
+    expect(result.summary.total).toBe(2);
+    expect(result.summary.globalViewAccess).toBe(1);
+    expect(result.summary.privateGroups).toBe(1);
+  });
+
+  it('should grant appropriate permissions during migration', async () => {
+    // Create prompt groups
+    const globalPromptGroup = await PromptGroup.create({
+      name: 'Global Group',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    const privatePromptGroup = await PromptGroup.create({
+      name: 'Private Group',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    // Add global group to project's promptGroupIds array
+    testProject.promptGroupIds = [globalPromptGroup._id];
+    await testProject.save();
+
+    const result = await migrateToPromptGroupPermissions({ dryRun: false });
+
+    expect(result.migrated).toBe(2);
+    expect(result.errors).toBe(0);
+    expect(result.ownerGrants).toBe(2);
+    expect(result.publicViewGrants).toBe(1);
+
+    // Check global promptGroup permissions
+    const globalOwnerEntry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: globalPromptGroup._id,
+      principalType: PrincipalType.USER,
+      principalId: testOwner._id,
+    });
+    expect(globalOwnerEntry).toBeTruthy();
+    expect(globalOwnerEntry.permBits).toBe(ownerRole.permBits);
+
+    const globalPublicEntry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: globalPromptGroup._id,
+      principalType: PrincipalType.PUBLIC,
+    });
+    expect(globalPublicEntry).toBeTruthy();
+    expect(globalPublicEntry.permBits).toBe(viewerRole.permBits);
+
+    // Check private promptGroup permissions
+    const privateOwnerEntry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: privatePromptGroup._id,
+      principalType: PrincipalType.USER,
+      principalId: testOwner._id,
+    });
+    expect(privateOwnerEntry).toBeTruthy();
+    expect(privateOwnerEntry.permBits).toBe(ownerRole.permBits);
+
+    const privatePublicEntry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: privatePromptGroup._id,
+      principalType: PrincipalType.PUBLIC,
+    });
+    expect(privatePublicEntry).toBeNull();
+  });
+
+  it('should skip promptGroups that already have ACL entries', async () => {
+    // Create prompt groups
+    const promptGroup1 = await PromptGroup.create({
+      name: 'Group 1',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    const promptGroup2 = await PromptGroup.create({
+      name: 'Group 2',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    // Grant permission to one promptGroup manually (simulating it already has ACL)
+    await AclEntry.create({
+      principalType: PrincipalType.USER,
+      principalId: testOwner._id,
+      principalModel: PrincipalModel.USER,
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: promptGroup1._id,
+      permBits: ownerRole.permBits,
+      roleId: ownerRole._id,
+      grantedBy: testOwner._id,
+      grantedAt: new Date(),
+    });
+
+    const result = await migrateToPromptGroupPermissions({ dryRun: false });
+
+    // Should only migrate promptGroup2, skip promptGroup1
+    expect(result.migrated).toBe(1);
+    expect(result.errors).toBe(0);
+
+    // Verify promptGroup2 now has permissions
+    const group2Entry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: promptGroup2._id,
+    });
+    expect(group2Entry).toBeTruthy();
+  });
+
+  it('should handle promptGroups with prompts correctly', async () => {
+    // Create a promptGroup with some prompts
+    const promptGroup = await PromptGroup.create({
+      name: 'Group with Prompts',
+      author: testOwner._id,
+      authorName: testOwner.name,
+      productionId: new ObjectId(),
+    });
+
+    // Create some prompts in this group
+    await Prompt.create({
+      prompt: 'First prompt',
+      author: testOwner._id,
+      groupId: promptGroup._id,
+      type: 'text',
+    });
+
+    await Prompt.create({
+      prompt: 'Second prompt',
+      author: testOwner._id,
+      groupId: promptGroup._id,
+      type: 'text',
+    });
+
+    const result = await migrateToPromptGroupPermissions({ dryRun: false });
+
+    expect(result.migrated).toBe(1);
+    expect(result.errors).toBe(0);
+
+    // Verify the promptGroup has permissions
+    const groupEntry = await AclEntry.findOne({
+      resourceType: ResourceType.PROMPTGROUP,
+      resourceId: promptGroup._id,
+    });
+    expect(groupEntry).toBeTruthy();
+
+    // Verify no prompt-level permissions were created
+    const promptEntries = await AclEntry.find({
+      resourceType: 'prompt',
+    });
+    expect(promptEntries).toHaveLength(0);
+  });
+});

api/models/Role.js

@@ -2,7 +2,6 @@ const {
   CacheKeys,
   SystemRoles,
   roleDefaults,
-  PermissionTypes,
   permissionsSchema,
   removeNullishValues,
 } = require('librechat-data-provider');
@@ -17,7 +16,7 @@ const { Role } = require('~/db/models');
  *
  * @param {string} roleName - The name of the role to find or create.
  * @param {string|string[]} [fieldsToSelect] - The fields to include or exclude in the returned document.
- * @returns {Promise<Object>} A plain object representing the role document.
+ * @returns {Promise<IRole>} Role document.
  */
 const getRoleByName = async function (roleName, fieldsToSelect = null) {
   const cache = getLogStores(CacheKeys.ROLES);
@@ -73,8 +72,9 @@ const updateRoleByName = async function (roleName, updates) {
  * Updates access permissions for a specific role and multiple permission types.
  * @param {string} roleName - The role to update.
  * @param {Object.<PermissionTypes, Object.<Permissions, boolean>>} permissionsUpdate - Permissions to update and their values.
+ * @param {IRole} [roleData] - Optional role data to use instead of fetching from the database.
  */
-async function updateAccessPermissions(roleName, permissionsUpdate) {
+async function updateAccessPermissions(roleName, permissionsUpdate, roleData) {
   // Filter and clean the permission updates based on our schema definition.
   const updates = {};
   for (const [permissionType, permissions] of Object.entries(permissionsUpdate)) {
@@ -87,7 +87,7 @@ async function updateAccessPermissions(roleName, permissionsUpdate) {
   }
 
   try {
-    const role = await getRoleByName(roleName);
+    const role = roleData ?? (await getRoleByName(roleName));
     if (!role) {
       return;
     }
@@ -114,7 +114,6 @@ async function updateAccessPermissions(roleName, permissionsUpdate) {
       }
     }
 
-    // Process the current updates
     for (const [permissionType, permissions] of Object.entries(updates)) {
       const currentTypePermissions = currentPermissions[permissionType] || {};
       updatedPermissions[permissionType] = { ...currentTypePermissions };

api/models/Token.js

@@ -1,42 +0,0 @@
-const { findToken, updateToken, createToken } = require('~/models');
-const { encryptV2 } = require('~/server/utils/crypto');
-
-/**
- * Handles the OAuth token by creating or updating the token.
- * @param {object} fields
- * @param {string} fields.userId - The user's ID.
- * @param {string} fields.token - The full token to store.
- * @param {string} fields.identifier - Unique, alternative identifier for the token.
- * @param {number} fields.expiresIn - The number of seconds until the token expires.
- * @param {object} fields.metadata - Additional metadata to store with the token.
- * @param {string} [fields.type="oauth"] - The type of token. Default is 'oauth'.
- */
-async function handleOAuthToken({
-  token,
-  userId,
-  identifier,
-  expiresIn,
-  metadata,
-  type = 'oauth',
-}) {
-  const encrypedToken = await encryptV2(token);
-  const tokenData = {
-    type,
-    userId,
-    metadata,
-    identifier,
-    token: encrypedToken,
-    expiresIn: parseInt(expiresIn, 10) || 3600,
-  };
-
-  const existingToken = await findToken({ userId, identifier });
-  if (existingToken) {
-    return await updateToken({ identifier }, tokenData);
-  } else {
-    return await createToken(tokenData);
-  }
-}
-
-module.exports = {
-  handleOAuthToken,
-};

api/models/Transaction.js

@@ -1,5 +1,4 @@
 const { logger } = require('@librechat/data-schemas');
-const { getBalanceConfig } = require('~/server/services/Config');
 const { getMultiplier, getCacheMultiplier } = require('./tx');
 const { Transaction, Balance } = require('~/db/models');
 
@@ -187,20 +186,23 @@ async function createAutoRefillTransaction(txData) {
 
 /**
  * Static method to create a transaction and update the balance
- * @param {txData} txData - Transaction data.
+ * @param {txData} _txData - Transaction data.
  */
-async function createTransaction(txData) {
+async function createTransaction(_txData) {
+  const { balance, transactions, ...txData } = _txData;
   if (txData.rawAmount != null && isNaN(txData.rawAmount)) {
     return;
   }
 
+  if (transactions?.enabled === false) {
+    return;
+  }
+
   const transaction = new Transaction(txData);
   transaction.endpointTokenConfig = txData.endpointTokenConfig;
   calculateTokenValue(transaction);
 
   await transaction.save();
-
-  const balance = await getBalanceConfig();
   if (!balance?.enabled) {
     return;
   }
@@ -221,9 +223,14 @@ async function createTransaction(txData) {
 
 /**
  * Static method to create a structured transaction and update the balance
- * @param {txData} txData - Transaction data.
+ * @param {txData} _txData - Transaction data.
  */
-async function createStructuredTransaction(txData) {
+async function createStructuredTransaction(_txData) {
+  const { balance, transactions, ...txData } = _txData;
+  if (transactions?.enabled === false) {
+    return;
+  }
+
   const transaction = new Transaction({
     ...txData,
     endpointTokenConfig: txData.endpointTokenConfig,
@@ -233,7 +240,6 @@ async function createStructuredTransaction(txData) {
 
   await transaction.save();
 
-  const balance = await getBalanceConfig();
   if (!balance?.enabled) {
     return;
   }

api/models/Transaction.spec.js

@@ -1,13 +1,9 @@
 const mongoose = require('mongoose');
 const { MongoMemoryServer } = require('mongodb-memory-server');
 const { spendTokens, spendStructuredTokens } = require('./spendTokens');
-const { getBalanceConfig } = require('~/server/services/Config');
 const { getMultiplier, getCacheMultiplier } = require('./tx');
-const { createTransaction } = require('./Transaction');
-const { Balance } = require('~/db/models');
-
-// Mock the custom config module so we can control the balance flag.
-jest.mock('~/server/services/Config');
+const { createTransaction, createStructuredTransaction } = require('./Transaction');
+const { Balance, Transaction } = require('~/db/models');
 
 let mongoServer;
 beforeAll(async () => {
@@ -23,8 +19,6 @@ afterAll(async () => {
 
 beforeEach(async () => {
   await mongoose.connection.dropDatabase();
-  // Default: enable balance updates in tests.
-  getBalanceConfig.mockResolvedValue({ enabled: true });
 });
 
 describe('Regular Token Spending Tests', () => {
@@ -41,6 +35,7 @@ describe('Regular Token Spending Tests', () => {
       model,
       context: 'test',
       endpointTokenConfig: null,
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -74,6 +69,7 @@ describe('Regular Token Spending Tests', () => {
       model,
       context: 'test',
       endpointTokenConfig: null,
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -104,6 +100,7 @@ describe('Regular Token Spending Tests', () => {
       model,
       context: 'test',
       endpointTokenConfig: null,
+      balance: { enabled: true },
     };
 
     const tokenUsage = {};
@@ -128,6 +125,7 @@ describe('Regular Token Spending Tests', () => {
       model,
       context: 'test',
       endpointTokenConfig: null,
+      balance: { enabled: true },
     };
 
     const tokenUsage = { promptTokens: 100 };
@@ -143,8 +141,7 @@ describe('Regular Token Spending Tests', () => {
   });
 
   test('spendTokens should not update balance when balance feature is disabled', async () => {
-    // Arrange: Override the config to disable balance updates.
-    getBalanceConfig.mockResolvedValue({ balance: { enabled: false } });
+    // Arrange: Balance config is now passed directly in txData
     const userId = new mongoose.Types.ObjectId();
     const initialBalance = 10000000;
     await Balance.create({ user: userId, tokenCredits: initialBalance });
@@ -156,6 +153,7 @@ describe('Regular Token Spending Tests', () => {
       model,
       context: 'test',
       endpointTokenConfig: null,
+      balance: { enabled: false },
     };
 
     const tokenUsage = {
@@ -186,6 +184,7 @@ describe('Structured Token Spending Tests', () => {
       model,
       context: 'message',
       endpointTokenConfig: null,
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -239,6 +238,7 @@ describe('Structured Token Spending Tests', () => {
       conversationId: 'test-convo',
       model,
       context: 'message',
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -271,6 +271,7 @@ describe('Structured Token Spending Tests', () => {
       conversationId: 'test-convo',
       model,
       context: 'message',
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -302,6 +303,7 @@ describe('Structured Token Spending Tests', () => {
       conversationId: 'test-convo',
       model,
       context: 'message',
+      balance: { enabled: true },
     };
 
     const tokenUsage = {};
@@ -328,6 +330,7 @@ describe('Structured Token Spending Tests', () => {
       conversationId: 'test-convo',
       model,
       context: 'incomplete',
+      balance: { enabled: true },
     };
 
     const tokenUsage = {
@@ -364,6 +367,7 @@ describe('NaN Handling Tests', () => {
       endpointTokenConfig: null,
       rawAmount: NaN,
       tokenType: 'prompt',
+      balance: { enabled: true },
     };
 
     // Act
@@ -375,3 +379,188 @@ describe('NaN Handling Tests', () => {
     expect(balance.tokenCredits).toBe(initialBalance);
   });
 });
+
+describe('Transactions Config Tests', () => {
+  test('createTransaction should not save when transactions.enabled is false', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'gpt-3.5-turbo';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'test',
+      endpointTokenConfig: null,
+      rawAmount: -100,
+      tokenType: 'prompt',
+      transactions: { enabled: false },
+    };
+
+    // Act
+    const result = await createTransaction(txData);
+
+    // Assert: No transaction should be created
+    expect(result).toBeUndefined();
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(0);
+    const balance = await Balance.findOne({ user: userId });
+    expect(balance.tokenCredits).toBe(initialBalance);
+  });
+
+  test('createTransaction should save when transactions.enabled is true', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'gpt-3.5-turbo';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'test',
+      endpointTokenConfig: null,
+      rawAmount: -100,
+      tokenType: 'prompt',
+      transactions: { enabled: true },
+      balance: { enabled: true },
+    };
+
+    // Act
+    const result = await createTransaction(txData);
+
+    // Assert: Transaction should be created
+    expect(result).toBeDefined();
+    expect(result.balance).toBeLessThan(initialBalance);
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(1);
+    expect(transactions[0].rawAmount).toBe(-100);
+  });
+
+  test('createTransaction should save when balance.enabled is true even if transactions config is missing', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'gpt-3.5-turbo';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'test',
+      endpointTokenConfig: null,
+      rawAmount: -100,
+      tokenType: 'prompt',
+      balance: { enabled: true },
+      // No transactions config provided
+    };
+
+    // Act
+    const result = await createTransaction(txData);
+
+    // Assert: Transaction should be created (backward compatibility)
+    expect(result).toBeDefined();
+    expect(result.balance).toBeLessThan(initialBalance);
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(1);
+  });
+
+  test('createTransaction should save transaction but not update balance when balance is disabled but transactions enabled', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'gpt-3.5-turbo';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'test',
+      endpointTokenConfig: null,
+      rawAmount: -100,
+      tokenType: 'prompt',
+      transactions: { enabled: true },
+      balance: { enabled: false },
+    };
+
+    // Act
+    const result = await createTransaction(txData);
+
+    // Assert: Transaction should be created but balance unchanged
+    expect(result).toBeUndefined();
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(1);
+    expect(transactions[0].rawAmount).toBe(-100);
+    const balance = await Balance.findOne({ user: userId });
+    expect(balance.tokenCredits).toBe(initialBalance);
+  });
+
+  test('createStructuredTransaction should not save when transactions.enabled is false', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'claude-3-5-sonnet';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'message',
+      tokenType: 'prompt',
+      inputTokens: -10,
+      writeTokens: -100,
+      readTokens: -5,
+      transactions: { enabled: false },
+    };
+
+    // Act
+    const result = await createStructuredTransaction(txData);
+
+    // Assert: No transaction should be created
+    expect(result).toBeUndefined();
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(0);
+    const balance = await Balance.findOne({ user: userId });
+    expect(balance.tokenCredits).toBe(initialBalance);
+  });
+
+  test('createStructuredTransaction should save transaction but not update balance when balance is disabled but transactions enabled', async () => {
+    // Arrange
+    const userId = new mongoose.Types.ObjectId();
+    const initialBalance = 10000000;
+    await Balance.create({ user: userId, tokenCredits: initialBalance });
+
+    const model = 'claude-3-5-sonnet';
+    const txData = {
+      user: userId,
+      conversationId: 'test-conversation-id',
+      model,
+      context: 'message',
+      tokenType: 'prompt',
+      inputTokens: -10,
+      writeTokens: -100,
+      readTokens: -5,
+      transactions: { enabled: true },
+      balance: { enabled: false },
+    };
+
+    // Act
+    const result = await createStructuredTransaction(txData);
+
+    // Assert: Transaction should be created but balance unchanged
+    expect(result).toBeUndefined();
+    const transactions = await Transaction.find({ user: userId });
+    expect(transactions).toHaveLength(1);
+    expect(transactions[0].inputTokens).toBe(-10);
+    expect(transactions[0].writeTokens).toBe(-100);
+    expect(transactions[0].readTokens).toBe(-5);
+    const balance = await Balance.findOne({ user: userId });
+    expect(balance.tokenCredits).toBe(initialBalance);
+  });
+});

api/models/balanceMethods.js

@@ -118,7 +118,7 @@ const addIntervalToDate = (date, value, unit) => {
  * @async
  * @function
  * @param {Object} params - The function parameters.
- * @param {Express.Request} params.req - The Express request object.
+ * @param {ServerRequest} params.req - The Express request object.
  * @param {Express.Response} params.res - The Express response object.
  * @param {Object} params.txData - The transaction data.
  * @param {string} params.txData.user - The user ID or identifier.

api/models/convoStructure.spec.js

@@ -1,47 +1,9 @@
 const mongoose = require('mongoose');
+const { buildTree } = require('librechat-data-provider');
 const { MongoMemoryServer } = require('mongodb-memory-server');
 const { getMessages, bulkSaveMessages } = require('./Message');
 const { Message } = require('~/db/models');
 
-// Original version of buildTree function
-function buildTree({ messages, fileMap }) {
-  if (messages === null) {
-    return null;
-  }
-
-  const messageMap = {};
-  const rootMessages = [];
-  const childrenCount = {};
-
-  messages.forEach((message) => {
-    const parentId = message.parentMessageId ?? '';
-    childrenCount[parentId] = (childrenCount[parentId] || 0) + 1;
-
-    const extendedMessage = {
-      ...message,
-      children: [],
-      depth: 0,
-      siblingIndex: childrenCount[parentId] - 1,
-    };
-
-    if (message.files && fileMap) {
-      extendedMessage.files = message.files.map((file) => fileMap[file.file_id ?? ''] ?? file);
-    }
-
-    messageMap[message.messageId] = extendedMessage;
-
-    const parentMessage = messageMap[parentId];
-    if (parentMessage) {
-      parentMessage.children.push(extendedMessage);
-      extendedMessage.depth = parentMessage.depth + 1;
-    } else {
-      rootMessages.push(extendedMessage);
-    }
-  });
-
-  return rootMessages;
-}
-
 let mongod;
 beforeAll(async () => {
   mongod = await MongoMemoryServer.create();

api/models/index.js

@@ -22,9 +22,17 @@ const {
 } = require('./Message');
 const { getConvoTitle, getConvo, saveConvo, deleteConvos } = require('./Conversation');
 const { getPreset, getPresets, savePreset, deletePresets } = require('./Preset');
+const { File } = require('~/db/models');
+
+const seedDatabase = async () => {
+  await methods.initializeRoles();
+  await methods.seedDefaultRoles();
+  await methods.ensureDefaultCategories();
+};
 
 module.exports = {
   ...methods,
+  seedDatabase,
   comparePassword,
   findFileById,
   createFile,
@@ -51,4 +59,6 @@ module.exports = {
   getPresets,
   savePreset,
   deletePresets,
+
+  Files: File,
 };

api/models/interface.js

@@ -0,0 +1,24 @@
+const { logger } = require('@librechat/data-schemas');
+const { updateInterfacePermissions: updateInterfacePerms } = require('@librechat/api');
+const { getRoleByName, updateAccessPermissions } = require('./Role');
+
+/**
+ * Update interface permissions based on app configuration.
+ * Must be done independently from loading the app config.
+ * @param {AppConfig} appConfig
+ */
+async function updateInterfacePermissions(appConfig) {
+  try {
+    await updateInterfacePerms({
+      appConfig,
+      getRoleByName,
+      updateAccessPermissions,
+    });
+  } catch (error) {
+    logger.error('Error updating interface permissions:', error);
+  }
+}
+
+module.exports = {
+  updateInterfacePermissions,
+};

api/models/inviteUser.js

@@ -1,6 +1,6 @@
 const mongoose = require('mongoose');
+const { getRandomValues } = require('@librechat/api');
 const { logger, hashToken } = require('@librechat/data-schemas');
-const { getRandomValues } = require('~/server/utils/crypto');
 const { createToken, findToken } = require('~/models');
 
 /**

api/models/spendTokens.js

@@ -5,13 +5,7 @@ const { createTransaction, createStructuredTransaction } = require('./Transactio
  *
  * @function
  * @async
- * @param {Object} txData - Transaction data.
- * @param {mongoose.Schema.Types.ObjectId} txData.user - The user ID.
- * @param {String} txData.conversationId - The ID of the conversation.
- * @param {String} txData.model - The model name.
- * @param {String} txData.context - The context in which the transaction is made.
- * @param {EndpointTokenConfig} [txData.endpointTokenConfig] - The current endpoint token config.
- * @param {String} [txData.valueKey] - The value key (optional).
+ * @param {txData} txData - Transaction data.
  * @param {Object} tokenUsage - The number of tokens used.
  * @param {Number} tokenUsage.promptTokens - The number of prompt tokens used.
  * @param {Number} tokenUsage.completionTokens - The number of completion tokens used.
@@ -69,13 +63,7 @@ const spendTokens = async (txData, tokenUsage) => {
  *
  * @function
  * @async
- * @param {Object} txData - Transaction data.
- * @param {mongoose.Schema.Types.ObjectId} txData.user - The user ID.
- * @param {String} txData.conversationId - The ID of the conversation.
- * @param {String} txData.model - The model name.
- * @param {String} txData.context - The context in which the transaction is made.
- * @param {EndpointTokenConfig} [txData.endpointTokenConfig] - The current endpoint token config.
- * @param {String} [txData.valueKey] - The value key (optional).
+ * @param {txData} txData - Transaction data.
  * @param {Object} tokenUsage - The number of tokens used.
  * @param {Object} tokenUsage.promptTokens - The number of prompt tokens used.
  * @param {Number} tokenUsage.promptTokens.input - The number of input tokens.

api/models/spendTokens.spec.js

@@ -5,7 +5,6 @@ const { createTransaction, createAutoRefillTransaction } = require('./Transactio
 
 require('~/db/models');
 
-// Mock the logger to prevent console output during tests
 jest.mock('~/config', () => ({
   logger: {
     debug: jest.fn(),
@@ -13,10 +12,6 @@ jest.mock('~/config', () => ({
   },
 }));
 
-// Mock the Config service
-const { getBalanceConfig } = require('~/server/services/Config');
-jest.mock('~/server/services/Config');
-
 describe('spendTokens', () => {
   let mongoServer;
   let userId;
@@ -44,8 +39,7 @@ describe('spendTokens', () => {
     // Create a new user ID for each test
     userId = new mongoose.Types.ObjectId();
 
-    // Mock the balance config to be enabled by default
-    getBalanceConfig.mockResolvedValue({ enabled: true });
+    // Balance config is now passed directly in txData
   });
 
   it('should create transactions for both prompt and completion tokens', async () => {
@@ -60,6 +54,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'gpt-3.5-turbo',
       context: 'test',
+      balance: { enabled: true },
     };
     const tokenUsage = {
       promptTokens: 100,
@@ -98,6 +93,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'gpt-3.5-turbo',
       context: 'test',
+      balance: { enabled: true },
     };
     const tokenUsage = {
       promptTokens: 100,
@@ -127,6 +123,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'gpt-3.5-turbo',
       context: 'test',
+      balance: { enabled: true },
     };
     const tokenUsage = {};
 
@@ -138,8 +135,7 @@ describe('spendTokens', () => {
   });
 
   it('should not update balance when the balance feature is disabled', async () => {
-    // Override configuration: disable balance updates
-    getBalanceConfig.mockResolvedValue({ enabled: false });
+    // Balance is now passed directly in txData
     // Create a balance for the user
     await Balance.create({
       user: userId,
@@ -151,6 +147,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'gpt-3.5-turbo',
       context: 'test',
+      balance: { enabled: false },
     };
     const tokenUsage = {
       promptTokens: 100,
@@ -180,6 +177,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'gpt-4', // Using a more expensive model
       context: 'test',
+      balance: { enabled: true },
     };
 
     // Spending more tokens than the user has balance for
@@ -233,6 +231,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo-1',
       model: 'gpt-4',
       context: 'test',
+      balance: { enabled: true },
     };
 
     const tokenUsage1 = {
@@ -252,6 +251,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo-2',
       model: 'gpt-4',
       context: 'test',
+      balance: { enabled: true },
     };
 
     const tokenUsage2 = {
@@ -292,6 +292,7 @@ describe('spendTokens', () => {
       tokenType: 'completion',
       rawAmount: -100,
       context: 'test',
+      balance: { enabled: true },
     });
 
     console.log('Direct Transaction.create result:', directResult);
@@ -316,6 +317,7 @@ describe('spendTokens', () => {
         conversationId: `test-convo-${model}`,
         model,
         context: 'test',
+        balance: { enabled: true },
       };
 
       const tokenUsage = {
@@ -352,6 +354,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo-1',
       model: 'claude-3-5-sonnet',
       context: 'test',
+      balance: { enabled: true },
     };
 
     const tokenUsage1 = {
@@ -375,6 +378,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo-2',
       model: 'claude-3-5-sonnet',
       context: 'test',
+      balance: { enabled: true },
     };
 
     const tokenUsage2 = {
@@ -426,6 +430,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'claude-3-5-sonnet', // Using a model that supports structured tokens
       context: 'test',
+      balance: { enabled: true },
     };
 
     // Spending more tokens than the user has balance for
@@ -505,6 +510,7 @@ describe('spendTokens', () => {
         conversationId,
         user: userId,
         model: usage.model,
+        balance: { enabled: true },
       };
 
       // Calculate expected spend for this transaction
@@ -617,6 +623,7 @@ describe('spendTokens', () => {
           tokenType: 'credits',
           context: 'concurrent-refill-test',
           rawAmount: refillAmount,
+          balance: { enabled: true },
         }),
       );
     }
@@ -683,6 +690,7 @@ describe('spendTokens', () => {
       conversationId: 'test-convo',
       model: 'claude-3-5-sonnet',
       context: 'test',
+      balance: { enabled: true },
     };
     const tokenUsage = {
       promptTokens: {

api/models/tx.js

@@ -1,4 +1,4 @@
-const { matchModelName } = require('../utils');
+const { matchModelName } = require('@librechat/api');
 const defaultRate = 6;
 
 /**
@@ -78,7 +78,7 @@ const tokenValues = Object.assign(
     'gpt-3.5-turbo-1106': { prompt: 1, completion: 2 },
     'o4-mini': { prompt: 1.1, completion: 4.4 },
     'o3-mini': { prompt: 1.1, completion: 4.4 },
-    o3: { prompt: 10, completion: 40 },
+    o3: { prompt: 2, completion: 8 },
     'o1-mini': { prompt: 1.1, completion: 4.4 },
     'o1-preview': { prompt: 15, completion: 60 },
     o1: { prompt: 15, completion: 60 },
@@ -87,6 +87,9 @@ const tokenValues = Object.assign(
     'gpt-4.1': { prompt: 2, completion: 8 },
     'gpt-4.5': { prompt: 75, completion: 150 },
     'gpt-4o-mini': { prompt: 0.15, completion: 0.6 },
+    'gpt-5': { prompt: 1.25, completion: 10 },
+    'gpt-5-mini': { prompt: 0.25, completion: 2 },
+    'gpt-5-nano': { prompt: 0.05, completion: 0.4 },
     'gpt-4o': { prompt: 2.5, completion: 10 },
     'gpt-4o-2024-05-13': { prompt: 5, completion: 15 },
     'gpt-4-1106': { prompt: 10, completion: 30 },
@@ -135,10 +138,11 @@ const tokenValues = Object.assign(
     'grok-2-1212': { prompt: 2.0, completion: 10.0 },
     'grok-2-latest': { prompt: 2.0, completion: 10.0 },
     'grok-2': { prompt: 2.0, completion: 10.0 },
-    'grok-3-mini-fast': { prompt: 0.4, completion: 4 },
+    'grok-3-mini-fast': { prompt: 0.6, completion: 4 },
     'grok-3-mini': { prompt: 0.3, completion: 0.5 },
     'grok-3-fast': { prompt: 5.0, completion: 25.0 },
     'grok-3': { prompt: 3.0, completion: 15.0 },
+    'grok-4': { prompt: 3.0, completion: 15.0 },
     'grok-beta': { prompt: 5.0, completion: 15.0 },
     'mistral-large': { prompt: 2.0, completion: 6.0 },
     'pixtral-large': { prompt: 2.0, completion: 6.0 },
@@ -146,6 +150,9 @@ const tokenValues = Object.assign(
     codestral: { prompt: 0.3, completion: 0.9 },
     'ministral-8b': { prompt: 0.1, completion: 0.1 },
     'ministral-3b': { prompt: 0.04, completion: 0.04 },
+    // GPT-OSS models
+    'gpt-oss-20b': { prompt: 0.05, completion: 0.2 },
+    'gpt-oss-120b': { prompt: 0.15, completion: 0.6 },
   },
   bedrockValues,
 );
@@ -213,6 +220,12 @@ const getValueKey = (model, endpoint) => {
     return 'gpt-4.1';
   } else if (modelName.includes('gpt-4o-2024-05-13')) {
     return 'gpt-4o-2024-05-13';
+  } else if (modelName.includes('gpt-5-nano')) {
+    return 'gpt-5-nano';
+  } else if (modelName.includes('gpt-5-mini')) {
+    return 'gpt-5-mini';
+  } else if (modelName.includes('gpt-5')) {
+    return 'gpt-5';
   } else if (modelName.includes('gpt-4o-mini')) {
     return 'gpt-4o-mini';
   } else if (modelName.includes('gpt-4o')) {

api/models/tx.spec.js

@@ -25,8 +25,14 @@ describe('getValueKey', () => {
     expect(getValueKey('gpt-4-some-other-info')).toBe('8k');
   });
 
-  it('should return undefined for model names that do not match any known patterns', () => {
-    expect(getValueKey('gpt-5-some-other-info')).toBeUndefined();
+  it('should return "gpt-5" for model name containing "gpt-5"', () => {
+    expect(getValueKey('gpt-5-some-other-info')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-2025-01-30')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-2025-01-30-0130')).toBe('gpt-5');
+    expect(getValueKey('openai/gpt-5')).toBe('gpt-5');
+    expect(getValueKey('openai/gpt-5-2025-01-30')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-turbo')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-0130')).toBe('gpt-5');
   });
 
   it('should return "gpt-3.5-turbo-1106" for model name containing "gpt-3.5-turbo-1106"', () => {
@@ -84,6 +90,29 @@ describe('getValueKey', () => {
     expect(getValueKey('gpt-4.1-nano-0125')).toBe('gpt-4.1-nano');
   });
 
+  it('should return "gpt-5" for model type of "gpt-5"', () => {
+    expect(getValueKey('gpt-5-2025-01-30')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-2025-01-30-0130')).toBe('gpt-5');
+    expect(getValueKey('openai/gpt-5')).toBe('gpt-5');
+    expect(getValueKey('openai/gpt-5-2025-01-30')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-turbo')).toBe('gpt-5');
+    expect(getValueKey('gpt-5-0130')).toBe('gpt-5');
+  });
+
+  it('should return "gpt-5-mini" for model type of "gpt-5-mini"', () => {
+    expect(getValueKey('gpt-5-mini-2025-01-30')).toBe('gpt-5-mini');
+    expect(getValueKey('openai/gpt-5-mini')).toBe('gpt-5-mini');
+    expect(getValueKey('gpt-5-mini-0130')).toBe('gpt-5-mini');
+    expect(getValueKey('gpt-5-mini-2025-01-30-0130')).toBe('gpt-5-mini');
+  });
+
+  it('should return "gpt-5-nano" for model type of "gpt-5-nano"', () => {
+    expect(getValueKey('gpt-5-nano-2025-01-30')).toBe('gpt-5-nano');
+    expect(getValueKey('openai/gpt-5-nano')).toBe('gpt-5-nano');
+    expect(getValueKey('gpt-5-nano-0130')).toBe('gpt-5-nano');
+    expect(getValueKey('gpt-5-nano-2025-01-30-0130')).toBe('gpt-5-nano');
+  });
+
   it('should return "gpt-4o" for model type of "gpt-4o"', () => {
     expect(getValueKey('gpt-4o-2024-08-06')).toBe('gpt-4o');
     expect(getValueKey('gpt-4o-2024-08-06-0718')).toBe('gpt-4o');
@@ -207,6 +236,48 @@ describe('getMultiplier', () => {
     );
   });
 
+  it('should return the correct multiplier for gpt-5', () => {
+    const valueKey = getValueKey('gpt-5-2025-01-30');
+    expect(getMultiplier({ valueKey, tokenType: 'prompt' })).toBe(tokenValues['gpt-5'].prompt);
+    expect(getMultiplier({ valueKey, tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5'].completion,
+    );
+    expect(getMultiplier({ model: 'gpt-5-preview', tokenType: 'prompt' })).toBe(
+      tokenValues['gpt-5'].prompt,
+    );
+    expect(getMultiplier({ model: 'openai/gpt-5', tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5'].completion,
+    );
+  });
+
+  it('should return the correct multiplier for gpt-5-mini', () => {
+    const valueKey = getValueKey('gpt-5-mini-2025-01-30');
+    expect(getMultiplier({ valueKey, tokenType: 'prompt' })).toBe(tokenValues['gpt-5-mini'].prompt);
+    expect(getMultiplier({ valueKey, tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5-mini'].completion,
+    );
+    expect(getMultiplier({ model: 'gpt-5-mini-preview', tokenType: 'prompt' })).toBe(
+      tokenValues['gpt-5-mini'].prompt,
+    );
+    expect(getMultiplier({ model: 'openai/gpt-5-mini', tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5-mini'].completion,
+    );
+  });
+
+  it('should return the correct multiplier for gpt-5-nano', () => {
+    const valueKey = getValueKey('gpt-5-nano-2025-01-30');
+    expect(getMultiplier({ valueKey, tokenType: 'prompt' })).toBe(tokenValues['gpt-5-nano'].prompt);
+    expect(getMultiplier({ valueKey, tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5-nano'].completion,
+    );
+    expect(getMultiplier({ model: 'gpt-5-nano-preview', tokenType: 'prompt' })).toBe(
+      tokenValues['gpt-5-nano'].prompt,
+    );
+    expect(getMultiplier({ model: 'openai/gpt-5-nano', tokenType: 'completion' })).toBe(
+      tokenValues['gpt-5-nano'].completion,
+    );
+  });
+
   it('should return the correct multiplier for gpt-4o', () => {
     const valueKey = getValueKey('gpt-4o-2024-08-06');
     expect(getMultiplier({ valueKey, tokenType: 'prompt' })).toBe(tokenValues['gpt-4o'].prompt);
@@ -307,10 +378,22 @@ describe('getMultiplier', () => {
   });
 
   it('should return defaultRate if derived valueKey does not match any known patterns', () => {
-    expect(getMultiplier({ tokenType: 'prompt', model: 'gpt-5-some-other-info' })).toBe(
+    expect(getMultiplier({ tokenType: 'prompt', model: 'gpt-10-some-other-info' })).toBe(
       defaultRate,
     );
   });
+
+  it('should return correct multipliers for GPT-OSS models', () => {
+    const models = ['gpt-oss-20b', 'gpt-oss-120b'];
+    models.forEach((key) => {
+      const expectedPrompt = tokenValues[key].prompt;
+      const expectedCompletion = tokenValues[key].completion;
+      expect(getMultiplier({ valueKey: key, tokenType: 'prompt' })).toBe(expectedPrompt);
+      expect(getMultiplier({ valueKey: key, tokenType: 'completion' })).toBe(expectedCompletion);
+      expect(getMultiplier({ model: key, tokenType: 'prompt' })).toBe(expectedPrompt);
+      expect(getMultiplier({ model: key, tokenType: 'completion' })).toBe(expectedCompletion);
+    });
+  });
 });
 
 describe('AWS Bedrock Model Tests', () => {
@@ -636,6 +719,15 @@ describe('Grok Model Tests - Pricing', () => {
       );
     });
 
+    test('should return correct prompt and completion rates for Grok 4 model', () => {
+      expect(getMultiplier({ model: 'grok-4-0709', tokenType: 'prompt' })).toBe(
+        tokenValues['grok-4'].prompt,
+      );
+      expect(getMultiplier({ model: 'grok-4-0709', tokenType: 'completion' })).toBe(
+        tokenValues['grok-4'].completion,
+      );
+    });
+
     test('should return correct prompt and completion rates for Grok 3 models with prefixes', () => {
       expect(getMultiplier({ model: 'xai/grok-3', tokenType: 'prompt' })).toBe(
         tokenValues['grok-3'].prompt,
@@ -662,6 +754,15 @@ describe('Grok Model Tests - Pricing', () => {
         tokenValues['grok-3-mini-fast'].completion,
       );
     });
+
+    test('should return correct prompt and completion rates for Grok 4 model with prefixes', () => {
+      expect(getMultiplier({ model: 'xai/grok-4-0709', tokenType: 'prompt' })).toBe(
+        tokenValues['grok-4'].prompt,
+      );
+      expect(getMultiplier({ model: 'xai/grok-4-0709', tokenType: 'completion' })).toBe(
+        tokenValues['grok-4'].completion,
+      );
+    });
   });
 });
 

api/models/userMethods.js

@@ -3,7 +3,7 @@ const bcrypt = require('bcryptjs');
 /**
  * Compares the provided password with the user's password.
  *
- * @param {MongoUser} user - The user to compare the password for.
+ * @param {IUser} user - The user to compare the password for.
  * @param {string} candidatePassword - The password to test against the user's password.
  * @returns {Promise<boolean>} A promise that resolves to a boolean indicating if the password matches.
  */

api/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@librechat/backend",
-  "version": "v0.7.8",
+  "version": "v0.8.0-rc4",
   "description": "",
   "scripts": {
     "start": "echo 'please run this from the root directory'",
@@ -34,30 +34,32 @@
   },
   "homepage": "https://librechat.ai",
   "dependencies": {
-    "@anthropic-ai/sdk": "^0.37.0",
+    "@anthropic-ai/sdk": "^0.52.0",
     "@aws-sdk/client-s3": "^3.758.0",
     "@aws-sdk/s3-request-presigner": "^3.758.0",
     "@azure/identity": "^4.7.0",
     "@azure/search-documents": "^12.0.0",
     "@azure/storage-blob": "^12.27.0",
-    "@google/generative-ai": "^0.23.0",
+    "@google/generative-ai": "^0.24.0",
     "@googleapis/youtube": "^20.0.0",
     "@keyv/redis": "^4.3.3",
-    "@langchain/community": "^0.3.44",
-    "@langchain/core": "^0.3.57",
-    "@langchain/google-genai": "^0.2.9",
-    "@langchain/google-vertexai": "^0.2.9",
+    "@langchain/community": "^0.3.47",
+    "@langchain/core": "^0.3.62",
+    "@langchain/google-genai": "^0.2.13",
+    "@langchain/google-vertexai": "^0.2.13",
+    "@langchain/openai": "^0.5.18",
     "@langchain/textsplitters": "^0.1.0",
-    "@librechat/agents": "^2.4.38",
+    "@librechat/agents": "^2.4.79",
     "@librechat/api": "*",
     "@librechat/data-schemas": "*",
-    "@node-saml/passport-saml": "^5.0.0",
+    "@microsoft/microsoft-graph-client": "^3.0.7",
+    "@modelcontextprotocol/sdk": "^1.17.1",
+    "@node-saml/passport-saml": "^5.1.0",
     "@waylaidwanderer/fetch-event-source": "^3.0.1",
-    "axios": "^1.8.2",
+    "axios": "^1.12.1",
     "bcryptjs": "^2.4.3",
-    "cohere-ai": "^7.9.1",
-    "compression": "^1.7.4",
-    "connect-redis": "^7.1.0",
+    "compression": "^1.8.1",
+    "connect-redis": "^8.1.0",
     "cookie": "^0.7.2",
     "cookie-parser": "^1.4.7",
     "cors": "^2.8.5",
@@ -67,10 +69,11 @@
     "express": "^4.21.2",
     "express-mongo-sanitize": "^2.2.0",
     "express-rate-limit": "^7.4.1",
-    "express-session": "^1.18.1",
+    "express-session": "^1.18.2",
     "express-static-gzip": "^2.2.0",
     "file-type": "^18.7.0",
     "firebase": "^11.0.2",
+    "form-data": "^4.0.4",
     "googleapis": "^126.0.1",
     "handlebars": "^4.7.7",
     "https-proxy-agent": "^7.0.6",
@@ -88,13 +91,12 @@
     "mime": "^3.0.0",
     "module-alias": "^2.2.3",
     "mongoose": "^8.12.1",
-    "multer": "^2.0.1",
+    "multer": "^2.0.2",
     "nanoid": "^3.3.7",
     "node-fetch": "^2.7.0",
     "nodemailer": "^6.9.15",
     "ollama": "^0.5.0",
-    "openai": "^4.96.2",
-    "openai-chat-tokens": "^0.2.8",
+    "openai": "^5.10.1",
     "openid-client": "^6.5.0",
     "passport": "^0.6.0",
     "passport-apple": "^2.0.2",
@@ -118,7 +120,7 @@
   },
   "devDependencies": {
     "jest": "^29.7.0",
-    "mongodb-memory-server": "^10.1.3",
+    "mongodb-memory-server": "^10.1.4",
     "nodemon": "^3.0.3",
     "supertest": "^7.1.0"
   }

api/server/cleanup.js

@@ -1,6 +1,6 @@
-const { logger } = require('~/config');
+const { logger } = require('@librechat/data-schemas');
 
-// WeakMap to hold temporary data associated with requests
+/** WeakMap to hold temporary data associated with requests */
 const requestDataMap = new WeakMap();
 
 const FinalizationRegistry = global.FinalizationRegistry || null;
@@ -23,7 +23,7 @@ const clientRegistry = FinalizationRegistry
         } else {
           logger.debug('[FinalizationRegistry] Cleaning up client');
         }
-      } catch (e) {
+      } catch {
         // Ignore errors
       }
     })
@@ -55,6 +55,9 @@ function disposeClient(client) {
     if (client.responseMessageId) {
       client.responseMessageId = null;
     }
+    if (client.parentMessageId) {
+      client.parentMessageId = null;
+    }
     if (client.message_file_map) {
       client.message_file_map = null;
     }
@@ -169,9 +172,6 @@ function disposeClient(client) {
       client.isGenerativeModel = null;
     }
     // Properties specific to OpenAIClient
-    if (client.ChatGPTClient) {
-      client.ChatGPTClient = null;
-    }
     if (client.completionsUrl) {
       client.completionsUrl = null;
     }
@@ -337,7 +337,7 @@ function disposeClient(client) {
       }
     }
     client.options = null;
-  } catch (e) {
+  } catch {
     // Ignore errors during disposal
   }
 }

api/server/controllers/AskController.js

@@ -1,282 +0,0 @@
-const { getResponseSender, Constants } = require('librechat-data-provider');
-const {
-  handleAbortError,
-  createAbortController,
-  cleanupAbortController,
-} = require('~/server/middleware');
-const {
-  disposeClient,
-  processReqData,
-  clientRegistry,
-  requestDataMap,
-} = require('~/server/cleanup');
-const { sendMessage, createOnProgress } = require('~/server/utils');
-const { saveMessage } = require('~/models');
-const { logger } = require('~/config');
-
-const AskController = async (req, res, next, initializeClient, addTitle) => {
-  let {
-    text,
-    endpointOption,
-    conversationId,
-    modelDisplayLabel,
-    parentMessageId = null,
-    overrideParentMessageId = null,
-  } = req.body;
-
-  let client = null;
-  let abortKey = null;
-  let cleanupHandlers = [];
-  let clientRef = null;
-
-  logger.debug('[AskController]', {
-    text,
-    conversationId,
-    ...endpointOption,
-    modelsConfig: endpointOption?.modelsConfig ? 'exists' : '',
-  });
-
-  let userMessage = null;
-  let userMessagePromise = null;
-  let promptTokens = null;
-  let userMessageId = null;
-  let responseMessageId = null;
-  let getAbortData = null;
-
-  const sender = getResponseSender({
-    ...endpointOption,
-    model: endpointOption.modelOptions.model,
-    modelDisplayLabel,
-  });
-  const initialConversationId = conversationId;
-  const newConvo = !initialConversationId;
-  const userId = req.user.id;
-
-  let reqDataContext = {
-    userMessage,
-    userMessagePromise,
-    responseMessageId,
-    promptTokens,
-    conversationId,
-    userMessageId,
-  };
-
-  const updateReqData = (data = {}) => {
-    reqDataContext = processReqData(data, reqDataContext);
-    abortKey = reqDataContext.abortKey;
-    userMessage = reqDataContext.userMessage;
-    userMessagePromise = reqDataContext.userMessagePromise;
-    responseMessageId = reqDataContext.responseMessageId;
-    promptTokens = reqDataContext.promptTokens;
-    conversationId = reqDataContext.conversationId;
-    userMessageId = reqDataContext.userMessageId;
-  };
-
-  let { onProgress: progressCallback, getPartialText } = createOnProgress();
-
-  const performCleanup = () => {
-    logger.debug('[AskController] Performing cleanup');
-    if (Array.isArray(cleanupHandlers)) {
-      for (const handler of cleanupHandlers) {
-        try {
-          if (typeof handler === 'function') {
-            handler();
-          }
-        } catch (e) {
-          // Ignore
-        }
-      }
-    }
-
-    if (abortKey) {
-      logger.debug('[AskController] Cleaning up abort controller');
-      cleanupAbortController(abortKey);
-      abortKey = null;
-    }
-
-    if (client) {
-      disposeClient(client);
-      client = null;
-    }
-
-    reqDataContext = null;
-    userMessage = null;
-    userMessagePromise = null;
-    promptTokens = null;
-    getAbortData = null;
-    progressCallback = null;
-    endpointOption = null;
-    cleanupHandlers = null;
-    addTitle = null;
-
-    if (requestDataMap.has(req)) {
-      requestDataMap.delete(req);
-    }
-    logger.debug('[AskController] Cleanup completed');
-  };
-
-  try {
-    ({ client } = await initializeClient({ req, res, endpointOption }));
-    if (clientRegistry && client) {
-      clientRegistry.register(client, { userId }, client);
-    }
-
-    if (client) {
-      requestDataMap.set(req, { client });
-    }
-
-    clientRef = new WeakRef(client);
-
-    getAbortData = () => {
-      const currentClient = clientRef?.deref();
-      const currentText =
-        currentClient?.getStreamText != null ? currentClient.getStreamText() : getPartialText();
-
-      return {
-        sender,
-        conversationId,
-        messageId: reqDataContext.responseMessageId,
-        parentMessageId: overrideParentMessageId ?? userMessageId,
-        text: currentText,
-        userMessage: userMessage,
-        userMessagePromise: userMessagePromise,
-        promptTokens: reqDataContext.promptTokens,
-      };
-    };
-
-    const { onStart, abortController } = createAbortController(
-      req,
-      res,
-      getAbortData,
-      updateReqData,
-    );
-
-    const closeHandler = () => {
-      logger.debug('[AskController] Request closed');
-      if (!abortController || abortController.signal.aborted || abortController.requestCompleted) {
-        return;
-      }
-      abortController.abort();
-      logger.debug('[AskController] Request aborted on close');
-    };
-
-    res.on('close', closeHandler);
-    cleanupHandlers.push(() => {
-      try {
-        res.removeListener('close', closeHandler);
-      } catch (e) {
-        // Ignore
-      }
-    });
-
-    const messageOptions = {
-      user: userId,
-      parentMessageId,
-      conversationId: reqDataContext.conversationId,
-      overrideParentMessageId,
-      getReqData: updateReqData,
-      onStart,
-      abortController,
-      progressCallback,
-      progressOptions: {
-        res,
-      },
-    };
-
-    /** @type {TMessage} */
-    let response = await client.sendMessage(text, messageOptions);
-    response.endpoint = endpointOption.endpoint;
-
-    const databasePromise = response.databasePromise;
-    delete response.databasePromise;
-
-    const { conversation: convoData = {} } = await databasePromise;
-    const conversation = { ...convoData };
-    conversation.title =
-      conversation && !conversation.title ? null : conversation?.title || 'New Chat';
-
-    const latestUserMessage = reqDataContext.userMessage;
-
-    if (client?.options?.attachments && latestUserMessage) {
-      latestUserMessage.files = client.options.attachments;
-      if (endpointOption?.modelOptions?.model) {
-        conversation.model = endpointOption.modelOptions.model;
-      }
-      delete latestUserMessage.image_urls;
-    }
-
-    if (!abortController.signal.aborted) {
-      const finalResponseMessage = { ...response };
-
-      sendMessage(res, {
-        final: true,
-        conversation,
-        title: conversation.title,
-        requestMessage: latestUserMessage,
-        responseMessage: finalResponseMessage,
-      });
-      res.end();
-
-      if (client?.savedMessageIds && !client.savedMessageIds.has(response.messageId)) {
-        await saveMessage(
-          req,
-          { ...finalResponseMessage, user: userId },
-          { context: 'api/server/controllers/AskController.js - response end' },
-        );
-      }
-    }
-
-    if (!client?.skipSaveUserMessage && latestUserMessage) {
-      await saveMessage(req, latestUserMessage, {
-        context: "api/server/controllers/AskController.js - don't skip saving user message",
-      });
-    }
-
-    if (typeof addTitle === 'function' && parentMessageId === Constants.NO_PARENT && newConvo) {
-      addTitle(req, {
-        text,
-        response: { ...response },
-        client,
-      })
-        .then(() => {
-          logger.debug('[AskController] Title generation started');
-        })
-        .catch((err) => {
-          logger.error('[AskController] Error in title generation', err);
-        })
-        .finally(() => {
-          logger.debug('[AskController] Title generation completed');
-          performCleanup();
-        });
-    } else {
-      performCleanup();
-    }
-  } catch (error) {
-    logger.error('[AskController] Error handling request', error);
-    let partialText = '';
-    try {
-      const currentClient = clientRef?.deref();
-      partialText =
-        currentClient?.getStreamText != null ? currentClient.getStreamText() : getPartialText();
-    } catch (getTextError) {
-      logger.error('[AskController] Error calling getText() during error handling', getTextError);
-    }
-
-    handleAbortError(res, req, error, {
-      sender,
-      partialText,
-      conversationId: reqDataContext.conversationId,
-      messageId: reqDataContext.responseMessageId,
-      parentMessageId: overrideParentMessageId ?? reqDataContext.userMessageId ?? parentMessageId,
-      userMessageId: reqDataContext.userMessageId,
-    })
-      .catch((err) => {
-        logger.error('[AskController] Error in `handleAbortError` during catch block', err);
-      })
-      .finally(() => {
-        performCleanup();
-      });
-  }
-};
-
-module.exports = AskController;

api/server/controllers/AuthController.js

@@ -1,17 +1,18 @@
 const cookies = require('cookie');
 const jwt = require('jsonwebtoken');
 const openIdClient = require('openid-client');
+const { isEnabled } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
 const {
-  registerUser,
-  resetPassword,
-  setAuthTokens,
   requestPasswordReset,
   setOpenIDAuthTokens,
+  resetPassword,
+  setAuthTokens,
+  registerUser,
 } = require('~/server/services/AuthService');
 const { findUser, getUserById, deleteAllUserSessions, findSession } = require('~/models');
 const { getOpenIdConfig } = require('~/strategies');
-const { isEnabled } = require('~/server/utils');
+const { getGraphApiToken } = require('~/server/services/GraphTokenService');
 
 const registrationController = async (req, res) => {
   try {
@@ -74,7 +75,7 @@ const refreshController = async (req, res) => {
       if (!user) {
         return res.status(401).redirect('/login');
       }
-      const token = setOpenIDAuthTokens(tokenset, res);
+      const token = setOpenIDAuthTokens(tokenset, res, user._id.toString());
       return res.status(200).send({ token, user });
     } catch (error) {
       logger.error('[refreshController] OpenID token refresh error', error);
@@ -83,7 +84,7 @@ const refreshController = async (req, res) => {
   }
   try {
     const payload = jwt.verify(refreshToken, process.env.JWT_REFRESH_SECRET);
-    const user = await getUserById(payload.id, '-password -__v -totpSecret');
+    const user = await getUserById(payload.id, '-password -__v -totpSecret -backupCodes');
     if (!user) {
       return res.status(401).redirect('/login');
     }
@@ -118,9 +119,54 @@ const refreshController = async (req, res) => {
   }
 };
 
+const graphTokenController = async (req, res) => {
+  try {
+    // Validate user is authenticated via Entra ID
+    if (!req.user.openidId || req.user.provider !== 'openid') {
+      return res.status(403).json({
+        message: 'Microsoft Graph access requires Entra ID authentication',
+      });
+    }
+
+    // Check if OpenID token reuse is active (required for on-behalf-of flow)
+    if (!isEnabled(process.env.OPENID_REUSE_TOKENS)) {
+      return res.status(403).json({
+        message: 'SharePoint integration requires OpenID token reuse to be enabled',
+      });
+    }
+
+    // Extract access token from Authorization header
+    const authHeader = req.headers.authorization;
+    if (!authHeader || !authHeader.startsWith('Bearer ')) {
+      return res.status(401).json({
+        message: 'Valid authorization token required',
+      });
+    }
+
+    // Get scopes from query parameters
+    const scopes = req.query.scopes;
+    if (!scopes) {
+      return res.status(400).json({
+        message: 'Graph API scopes are required as query parameter',
+      });
+    }
+
+    const accessToken = authHeader.substring(7); // Remove 'Bearer ' prefix
+    const tokenResponse = await getGraphApiToken(req.user, accessToken, scopes);
+
+    res.json(tokenResponse);
+  } catch (error) {
+    logger.error('[graphTokenController] Failed to obtain Graph API token:', error);
+    res.status(500).json({
+      message: 'Failed to obtain Microsoft Graph token',
+    });
+  }
+};
+
 module.exports = {
   refreshController,
   registrationController,
   resetPasswordController,
   resetPasswordRequestController,
+  graphTokenController,
 };

api/server/controllers/EditController.js

@@ -1,3 +1,5 @@
+const { sendEvent } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
 const { getResponseSender } = require('librechat-data-provider');
 const {
   handleAbortError,
@@ -10,9 +12,8 @@ const {
   clientRegistry,
   requestDataMap,
 } = require('~/server/cleanup');
-const { sendMessage, createOnProgress } = require('~/server/utils');
+const { createOnProgress } = require('~/server/utils');
 const { saveMessage } = require('~/models');
-const { logger } = require('~/config');
 
 const EditController = async (req, res, next, initializeClient) => {
   let {
@@ -84,7 +85,7 @@ const EditController = async (req, res, next, initializeClient) => {
     }
 
     if (abortKey) {
-      logger.debug('[AskController] Cleaning up abort controller');
+      logger.debug('[EditController] Cleaning up abort controller');
       cleanupAbortController(abortKey);
       abortKey = null;
     }
@@ -198,7 +199,7 @@ const EditController = async (req, res, next, initializeClient) => {
       const finalUserMessage = reqDataContext.userMessage;
       const finalResponseMessage = { ...response };
 
-      sendMessage(res, {
+      sendEvent(res, {
         final: true,
         conversation,
         title: conversation.title,

api/server/controllers/ErrorController.js

@@ -1,40 +0,0 @@
-const { logger } = require('~/config');
-
-//handle duplicates
-const handleDuplicateKeyError = (err, res) => {
-  logger.error('Duplicate key error:', err.keyValue);
-  const field = `${JSON.stringify(Object.keys(err.keyValue))}`;
-  const code = 409;
-  res
-    .status(code)
-    .send({ messages: `An document with that ${field} already exists.`, fields: field });
-};
-
-//handle validation errors
-const handleValidationError = (err, res) => {
-  logger.error('Validation error:', err.errors);
-  let errors = Object.values(err.errors).map((el) => el.message);
-  let fields = `${JSON.stringify(Object.values(err.errors).map((el) => el.path))}`;
-  let code = 400;
-  if (errors.length > 1) {
-    errors = errors.join(' ');
-    res.status(code).send({ messages: `${JSON.stringify(errors)}`, fields: fields });
-  } else {
-    res.status(code).send({ messages: `${JSON.stringify(errors)}`, fields: fields });
-  }
-};
-
-// eslint-disable-next-line no-unused-vars
-module.exports = (err, req, res, next) => {
-  try {
-    if (err.name === 'ValidationError') {
-      return (err = handleValidationError(err, res));
-    }
-    if (err.code && err.code == 11000) {
-      return (err = handleDuplicateKeyError(err, res));
-    }
-  } catch (err) {
-    logger.error('ErrorController => error', err);
-    res.status(500).send('An unknown error occurred.');
-  }
-};

api/server/controllers/ModelController.js

@@ -5,6 +5,7 @@ const { logger } = require('~/config');
 
 /**
  * @param {ServerRequest} req
+ * @returns {Promise<TModelsConfig>} The models config.
  */
 const getModelsConfig = async (req) => {
   const cache = getLogStores(CacheKeys.CONFIG_STORE);

api/server/controllers/OverrideController.js

@@ -1,27 +0,0 @@
-const { CacheKeys } = require('librechat-data-provider');
-const { loadOverrideConfig } = require('~/server/services/Config');
-const { getLogStores } = require('~/cache');
-
-async function overrideController(req, res) {
-  const cache = getLogStores(CacheKeys.CONFIG_STORE);
-  let overrideConfig = await cache.get(CacheKeys.OVERRIDE_CONFIG);
-  if (overrideConfig) {
-    res.send(overrideConfig);
-    return;
-  } else if (overrideConfig === false) {
-    res.send(false);
-    return;
-  }
-  overrideConfig = await loadOverrideConfig();
-  const { endpointsConfig, modelsConfig } = overrideConfig;
-  if (endpointsConfig) {
-    await cache.set(CacheKeys.ENDPOINT_CONFIG, endpointsConfig);
-  }
-  if (modelsConfig) {
-    await cache.set(CacheKeys.MODELS_CONFIG, modelsConfig);
-  }
-  await cache.set(CacheKeys.OVERRIDE_CONFIG, overrideConfig);
-  res.send(JSON.stringify(overrideConfig));
-}
-
-module.exports = overrideController;

api/server/controllers/PermissionsController.js

@@ -0,0 +1,484 @@
+/**
+ * @import { TUpdateResourcePermissionsRequest, TUpdateResourcePermissionsResponse } from 'librechat-data-provider'
+ */
+
+const mongoose = require('mongoose');
+const { logger } = require('@librechat/data-schemas');
+const { ResourceType, PrincipalType } = require('librechat-data-provider');
+const {
+  bulkUpdateResourcePermissions,
+  ensureGroupPrincipalExists,
+  getEffectivePermissions,
+  ensurePrincipalExists,
+  getAvailableRoles,
+} = require('~/server/services/PermissionService');
+const { AclEntry } = require('~/db/models');
+const {
+  searchPrincipals: searchLocalPrincipals,
+  sortPrincipalsByRelevance,
+  calculateRelevanceScore,
+} = require('~/models');
+const {
+  entraIdPrincipalFeatureEnabled,
+  searchEntraIdPrincipals,
+} = require('~/server/services/GraphApiService');
+
+/**
+ * Generic controller for resource permission endpoints
+ * Delegates validation and logic to PermissionService
+ */
+
+/**
+ * Validates that the resourceType is one of the supported enum values
+ * @param {string} resourceType - The resource type to validate
+ * @throws {Error} If resourceType is not valid
+ */
+const validateResourceType = (resourceType) => {
+  const validTypes = Object.values(ResourceType);
+  if (!validTypes.includes(resourceType)) {
+    throw new Error(`Invalid resourceType: ${resourceType}. Valid types: ${validTypes.join(', ')}`);
+  }
+};
+
+/**
+ * Bulk update permissions for a resource (grant, update, remove)
+ * @route PUT /api/{resourceType}/{resourceId}/permissions
+ * @param {Object} req - Express request object
+ * @param {Object} req.params - Route parameters
+ * @param {string} req.params.resourceType - Resource type (e.g., 'agent')
+ * @param {string} req.params.resourceId - Resource ID
+ * @param {TUpdateResourcePermissionsRequest} req.body - Request body
+ * @param {Object} res - Express response object
+ * @returns {Promise<TUpdateResourcePermissionsResponse>} Updated permissions response
+ */
+const updateResourcePermissions = async (req, res) => {
+  try {
+    const { resourceType, resourceId } = req.params;
+    validateResourceType(resourceType);
+
+    /** @type {TUpdateResourcePermissionsRequest} */
+    const { updated, removed, public: isPublic, publicAccessRoleId } = req.body;
+    const { id: userId } = req.user;
+
+    // Prepare principals for the service call
+    const updatedPrincipals = [];
+    const revokedPrincipals = [];
+
+    // Add updated principals
+    if (updated && Array.isArray(updated)) {
+      updatedPrincipals.push(...updated);
+    }
+
+    // Add public permission if enabled
+    if (isPublic && publicAccessRoleId) {
+      updatedPrincipals.push({
+        type: PrincipalType.PUBLIC,
+        id: null,
+        accessRoleId: publicAccessRoleId,
+      });
+    }
+
+    // Prepare authentication context for enhanced group member fetching
+    const useEntraId = entraIdPrincipalFeatureEnabled(req.user);
+    const authHeader = req.headers.authorization;
+    const accessToken =
+      authHeader && authHeader.startsWith('Bearer ') ? authHeader.substring(7) : null;
+    const authContext =
+      useEntraId && accessToken
+        ? {
+            accessToken,
+            sub: req.user.openidId,
+          }
+        : null;
+
+    // Ensure updated principals exist in the database before processing permissions
+    const validatedPrincipals = [];
+    for (const principal of updatedPrincipals) {
+      try {
+        let principalId;
+
+        if (principal.type === PrincipalType.PUBLIC) {
+          principalId = null; // Public principals don't need database records
+        } else if (principal.type === PrincipalType.ROLE) {
+          principalId = principal.id; // Role principals use role name as ID
+        } else if (principal.type === PrincipalType.USER) {
+          principalId = await ensurePrincipalExists(principal);
+        } else if (principal.type === PrincipalType.GROUP) {
+          // Pass authContext to enable member fetching for Entra ID groups when available
+          principalId = await ensureGroupPrincipalExists(principal, authContext);
+        } else {
+          logger.error(`Unsupported principal type: ${principal.type}`);
+          continue; // Skip invalid principal types
+        }
+
+        // Update the principal with the validated ID for ACL operations
+        validatedPrincipals.push({
+          ...principal,
+          id: principalId,
+        });
+      } catch (error) {
+        logger.error('Error ensuring principal exists:', {
+          principal: {
+            type: principal.type,
+            id: principal.id,
+            name: principal.name,
+            source: principal.source,
+          },
+          error: error.message,
+        });
+        // Continue with other principals instead of failing the entire operation
+        continue;
+      }
+    }
+
+    // Add removed principals
+    if (removed && Array.isArray(removed)) {
+      revokedPrincipals.push(...removed);
+    }
+
+    // If public is disabled, add public to revoked list
+    if (!isPublic) {
+      revokedPrincipals.push({
+        type: PrincipalType.PUBLIC,
+        id: null,
+      });
+    }
+
+    const results = await bulkUpdateResourcePermissions({
+      resourceType,
+      resourceId,
+      updatedPrincipals: validatedPrincipals,
+      revokedPrincipals,
+      grantedBy: userId,
+    });
+
+    /** @type {TUpdateResourcePermissionsResponse} */
+    const response = {
+      message: 'Permissions updated successfully',
+      results: {
+        principals: results.granted,
+        public: isPublic || false,
+        publicAccessRoleId: isPublic ? publicAccessRoleId : undefined,
+      },
+    };
+
+    res.status(200).json(response);
+  } catch (error) {
+    logger.error('Error updating resource permissions:', error);
+    res.status(400).json({
+      error: 'Failed to update permissions',
+      details: error.message,
+    });
+  }
+};
+
+/**
+ * Get principals with their permission roles for a resource (UI-friendly format)
+ * Uses efficient aggregation pipeline to join User/Group data in single query
+ * @route GET /api/permissions/{resourceType}/{resourceId}
+ */
+const getResourcePermissions = async (req, res) => {
+  try {
+    const { resourceType, resourceId } = req.params;
+    validateResourceType(resourceType);
+
+    // Use aggregation pipeline for efficient single-query data retrieval
+    const results = await AclEntry.aggregate([
+      // Match ACL entries for this resource
+      {
+        $match: {
+          resourceType,
+          resourceId: mongoose.Types.ObjectId.isValid(resourceId)
+            ? mongoose.Types.ObjectId.createFromHexString(resourceId)
+            : resourceId,
+        },
+      },
+      // Lookup AccessRole information
+      {
+        $lookup: {
+          from: 'accessroles',
+          localField: 'roleId',
+          foreignField: '_id',
+          as: 'role',
+        },
+      },
+      // Lookup User information (for user principals)
+      {
+        $lookup: {
+          from: 'users',
+          localField: 'principalId',
+          foreignField: '_id',
+          as: 'userInfo',
+        },
+      },
+      // Lookup Group information (for group principals)
+      {
+        $lookup: {
+          from: 'groups',
+          localField: 'principalId',
+          foreignField: '_id',
+          as: 'groupInfo',
+        },
+      },
+      // Project final structure
+      {
+        $project: {
+          principalType: 1,
+          principalId: 1,
+          accessRoleId: { $arrayElemAt: ['$role.accessRoleId', 0] },
+          userInfo: { $arrayElemAt: ['$userInfo', 0] },
+          groupInfo: { $arrayElemAt: ['$groupInfo', 0] },
+        },
+      },
+    ]);
+
+    const principals = [];
+    let publicPermission = null;
+
+    // Process aggregation results
+    for (const result of results) {
+      if (result.principalType === PrincipalType.PUBLIC) {
+        publicPermission = {
+          public: true,
+          publicAccessRoleId: result.accessRoleId,
+        };
+      } else if (result.principalType === PrincipalType.USER && result.userInfo) {
+        principals.push({
+          type: PrincipalType.USER,
+          id: result.userInfo._id.toString(),
+          name: result.userInfo.name || result.userInfo.username,
+          email: result.userInfo.email,
+          avatar: result.userInfo.avatar,
+          source: !result.userInfo._id ? 'entra' : 'local',
+          idOnTheSource: result.userInfo.idOnTheSource || result.userInfo._id.toString(),
+          accessRoleId: result.accessRoleId,
+        });
+      } else if (result.principalType === PrincipalType.GROUP && result.groupInfo) {
+        principals.push({
+          type: PrincipalType.GROUP,
+          id: result.groupInfo._id.toString(),
+          name: result.groupInfo.name,
+          email: result.groupInfo.email,
+          description: result.groupInfo.description,
+          avatar: result.groupInfo.avatar,
+          source: result.groupInfo.source || 'local',
+          idOnTheSource: result.groupInfo.idOnTheSource || result.groupInfo._id.toString(),
+          accessRoleId: result.accessRoleId,
+        });
+      } else if (result.principalType === PrincipalType.ROLE) {
+        principals.push({
+          type: PrincipalType.ROLE,
+          /** Role name as ID */
+          id: result.principalId,
+          /** Display the role name */
+          name: result.principalId,
+          description: `System role: ${result.principalId}`,
+          accessRoleId: result.accessRoleId,
+        });
+      }
+    }
+
+    // Return response in format expected by frontend
+    const response = {
+      resourceType,
+      resourceId,
+      principals,
+      public: publicPermission?.public || false,
+      ...(publicPermission?.publicAccessRoleId && {
+        publicAccessRoleId: publicPermission.publicAccessRoleId,
+      }),
+    };
+
+    res.status(200).json(response);
+  } catch (error) {
+    logger.error('Error getting resource permissions principals:', error);
+    res.status(500).json({
+      error: 'Failed to get permissions principals',
+      details: error.message,
+    });
+  }
+};
+
+/**
+ * Get available roles for a resource type
+ * @route GET /api/{resourceType}/roles
+ */
+const getResourceRoles = async (req, res) => {
+  try {
+    const { resourceType } = req.params;
+    validateResourceType(resourceType);
+
+    const roles = await getAvailableRoles({ resourceType });
+
+    res.status(200).json(
+      roles.map((role) => ({
+        accessRoleId: role.accessRoleId,
+        name: role.name,
+        description: role.description,
+        permBits: role.permBits,
+      })),
+    );
+  } catch (error) {
+    logger.error('Error getting resource roles:', error);
+    res.status(500).json({
+      error: 'Failed to get roles',
+      details: error.message,
+    });
+  }
+};
+
+/**
+ * Get user's effective permission bitmask for a resource
+ * @route GET /api/{resourceType}/{resourceId}/effective
+ */
+const getUserEffectivePermissions = async (req, res) => {
+  try {
+    const { resourceType, resourceId } = req.params;
+    validateResourceType(resourceType);
+
+    const { id: userId } = req.user;
+
+    const permissionBits = await getEffectivePermissions({
+      userId,
+      role: req.user.role,
+      resourceType,
+      resourceId,
+    });
+
+    res.status(200).json({
+      permissionBits,
+    });
+  } catch (error) {
+    logger.error('Error getting user effective permissions:', error);
+    res.status(500).json({
+      error: 'Failed to get effective permissions',
+      details: error.message,
+    });
+  }
+};
+
+/**
+ * Search for users and groups to grant permissions
+ * Supports hybrid local database + Entra ID search when configured
+ * @route GET /api/permissions/search-principals
+ */
+const searchPrincipals = async (req, res) => {
+  try {
+    const { q: query, limit = 20, types } = req.query;
+
+    if (!query || query.trim().length === 0) {
+      return res.status(400).json({
+        error: 'Query parameter "q" is required and must not be empty',
+      });
+    }
+
+    if (query.trim().length < 2) {
+      return res.status(400).json({
+        error: 'Query must be at least 2 characters long',
+      });
+    }
+
+    const searchLimit = Math.min(Math.max(1, parseInt(limit) || 10), 50);
+
+    let typeFilters = null;
+    if (types) {
+      const typesArray = Array.isArray(types) ? types : types.split(',');
+      const validTypes = typesArray.filter((t) =>
+        [PrincipalType.USER, PrincipalType.GROUP, PrincipalType.ROLE].includes(t),
+      );
+      typeFilters = validTypes.length > 0 ? validTypes : null;
+    }
+
+    const localResults = await searchLocalPrincipals(query.trim(), searchLimit, typeFilters);
+    let allPrincipals = [...localResults];
+
+    const useEntraId = entraIdPrincipalFeatureEnabled(req.user);
+
+    if (useEntraId && localResults.length < searchLimit) {
+      try {
+        let graphType = 'all';
+        if (typeFilters && typeFilters.length === 1) {
+          const graphTypeMap = {
+            [PrincipalType.USER]: 'users',
+            [PrincipalType.GROUP]: 'groups',
+          };
+          const mappedType = graphTypeMap[typeFilters[0]];
+          if (mappedType) {
+            graphType = mappedType;
+          }
+        }
+
+        const authHeader = req.headers.authorization;
+        const accessToken =
+          authHeader && authHeader.startsWith('Bearer ') ? authHeader.substring(7) : null;
+
+        if (accessToken) {
+          const graphResults = await searchEntraIdPrincipals(
+            accessToken,
+            req.user.openidId,
+            query.trim(),
+            graphType,
+            searchLimit - localResults.length,
+          );
+
+          const localEmails = new Set(
+            localResults.map((p) => p.email?.toLowerCase()).filter(Boolean),
+          );
+          const localGroupSourceIds = new Set(
+            localResults.map((p) => p.idOnTheSource).filter(Boolean),
+          );
+
+          for (const principal of graphResults) {
+            const isDuplicateByEmail =
+              principal.email && localEmails.has(principal.email.toLowerCase());
+            const isDuplicateBySourceId =
+              principal.idOnTheSource && localGroupSourceIds.has(principal.idOnTheSource);
+
+            if (!isDuplicateByEmail && !isDuplicateBySourceId) {
+              allPrincipals.push(principal);
+            }
+          }
+        }
+      } catch (graphError) {
+        logger.warn('Graph API search failed, falling back to local results:', graphError.message);
+      }
+    }
+    const scoredResults = allPrincipals.map((item) => ({
+      ...item,
+      _searchScore: calculateRelevanceScore(item, query.trim()),
+    }));
+
+    const finalResults = sortPrincipalsByRelevance(scoredResults)
+      .slice(0, searchLimit)
+      .map((result) => {
+        const { _searchScore, ...resultWithoutScore } = result;
+        return resultWithoutScore;
+      });
+
+    res.status(200).json({
+      query: query.trim(),
+      limit: searchLimit,
+      types: typeFilters,
+      results: finalResults,
+      count: finalResults.length,
+      sources: {
+        local: finalResults.filter((r) => r.source === 'local').length,
+        entra: finalResults.filter((r) => r.source === 'entra').length,
+      },
+    });
+  } catch (error) {
+    logger.error('Error searching principals:', error);
+    res.status(500).json({
+      error: 'Failed to search principals',
+      details: error.message,
+    });
+  }
+};
+
+module.exports = {
+  updateResourcePermissions,
+  getResourcePermissions,
+  getResourceRoles,
+  getUserEffectivePermissions,
+  searchPrincipals,
+};

api/server/controllers/PluginController.js

@@ -1,53 +1,18 @@
-const { CacheKeys, AuthType } = require('librechat-data-provider');
-const { getToolkitKey } = require('~/server/services/ToolService');
-const { getCustomConfig } = require('~/server/services/Config');
-const { availableTools } = require('~/app/clients/tools');
+const { logger } = require('@librechat/data-schemas');
+const { CacheKeys, Constants } = require('librechat-data-provider');
+const {
+  getToolkitKey,
+  checkPluginAuth,
+  filterUniquePlugins,
+  convertMCPToolToPlugin,
+  convertMCPToolsToPlugins,
+} = require('@librechat/api');
+const { getCachedTools, setCachedTools, mergeUserTools } = require('~/server/services/Config');
+const { availableTools, toolkits } = require('~/app/clients/tools');
+const { getAppConfig } = require('~/server/services/Config');
 const { getMCPManager } = require('~/config');
 const { getLogStores } = require('~/cache');
 
-/**
- * Filters out duplicate plugins from the list of plugins.
- *
- * @param {TPlugin[]} plugins The list of plugins to filter.
- * @returns {TPlugin[]} The list of plugins with duplicates removed.
- */
-const filterUniquePlugins = (plugins) => {
-  const seen = new Set();
-  return plugins.filter((plugin) => {
-    const duplicate = seen.has(plugin.pluginKey);
-    seen.add(plugin.pluginKey);
-    return !duplicate;
-  });
-};
-
-/**
- * Determines if a plugin is authenticated by checking if all required authentication fields have non-empty values.
- * Supports alternate authentication fields, allowing validation against multiple possible environment variables.
- *
- * @param {TPlugin} plugin The plugin object containing the authentication configuration.
- * @returns {boolean} True if the plugin is authenticated for all required fields, false otherwise.
- */
-const checkPluginAuth = (plugin) => {
-  if (!plugin.authConfig || plugin.authConfig.length === 0) {
-    return false;
-  }
-
-  return plugin.authConfig.every((authFieldObj) => {
-    const authFieldOptions = authFieldObj.authField.split('||');
-    let isFieldAuthenticated = false;
-
-    for (const fieldOption of authFieldOptions) {
-      const envValue = process.env[fieldOption];
-      if (envValue && envValue.trim() !== '' && envValue !== AuthType.USER_PROVIDED) {
-        isFieldAuthenticated = true;
-        break;
-      }
-    }
-
-    return isFieldAuthenticated;
-  });
-};
-
 const getAvailablePluginsController = async (req, res) => {
   try {
     const cache = getLogStores(CacheKeys.CONFIG_STORE);
@@ -57,8 +22,10 @@ const getAvailablePluginsController = async (req, res) => {
       return;
     }
 
+    const appConfig = await getAppConfig({ role: req.user?.role });
     /** @type {{ filteredTools: string[], includedTools: string[] }} */
-    const { filteredTools = [], includedTools = [] } = req.app.locals;
+    const { filteredTools = [], includedTools = [] } = appConfig;
+    /** @type {import('@librechat/api').LCManifestTool[]} */
     const pluginManifest = availableTools;
 
     const uniquePlugins = filterUniquePlugins(pluginManifest);
@@ -98,23 +65,72 @@ const getAvailablePluginsController = async (req, res) => {
  */
 const getAvailableTools = async (req, res) => {
   try {
+    const userId = req.user?.id;
+    if (!userId) {
+      logger.warn('[getAvailableTools] User ID not found in request');
+      return res.status(401).json({ message: 'Unauthorized' });
+    }
     const cache = getLogStores(CacheKeys.CONFIG_STORE);
-    const cachedTools = await cache.get(CacheKeys.TOOLS);
-    if (cachedTools) {
-      res.status(200).json(cachedTools);
+    const cachedToolsArray = await cache.get(CacheKeys.TOOLS);
+    const cachedUserTools = await getCachedTools({ userId });
+
+    const appConfig = req.config ?? (await getAppConfig({ role: req.user?.role }));
+
+    /** @type {TPlugin[]} */
+    let mcpPlugins;
+    if (appConfig?.mcpConfig) {
+      const mcpManager = getMCPManager();
+      mcpPlugins =
+        cachedUserTools != null
+          ? convertMCPToolsToPlugins({ functionTools: cachedUserTools, mcpManager })
+          : undefined;
+    }
+
+    if (
+      cachedToolsArray != null &&
+      (appConfig?.mcpConfig != null ? mcpPlugins != null && mcpPlugins.length > 0 : true)
+    ) {
+      const dedupedTools = filterUniquePlugins([...(mcpPlugins ?? []), ...cachedToolsArray]);
+      res.status(200).json(dedupedTools);
       return;
     }
 
+    /** @type {Record<string, FunctionTool> | null} Get tool definitions to filter which tools are actually available */
+    let toolDefinitions = await getCachedTools({ includeGlobal: true });
+    let prelimCachedTools;
+
+    /** @type {import('@librechat/api').LCManifestTool[]} */
     let pluginManifest = availableTools;
-    const customConfig = await getCustomConfig();
-    if (customConfig?.mcpServers != null) {
-      const mcpManager = getMCPManager();
-      pluginManifest = await mcpManager.loadManifestTools(pluginManifest);
+
+    if (appConfig?.mcpConfig != null) {
+      try {
+        const mcpManager = getMCPManager();
+        const mcpTools = await mcpManager.getAllToolFunctions(userId);
+        prelimCachedTools = prelimCachedTools ?? {};
+        for (const [toolKey, toolData] of Object.entries(mcpTools)) {
+          const plugin = convertMCPToolToPlugin({
+            toolKey,
+            toolData,
+            mcpManager,
+          });
+          if (plugin) {
+            pluginManifest.push(plugin);
+          }
+          prelimCachedTools[toolKey] = toolData;
+        }
+        await mergeUserTools({ userId, cachedUserTools, userTools: prelimCachedTools });
+      } catch (error) {
+        logger.error(
+          '[getAvailableTools] Error loading MCP Tools, servers may still be initializing:',
+          error,
+        );
+      }
+    } else if (prelimCachedTools != null) {
+      await setCachedTools(prelimCachedTools, { isGlobal: true });
     }
 
-    /** @type {TPlugin[]} */
+    /** @type {TPlugin[]} Deduplicate and authenticate plugins */
     const uniquePlugins = filterUniquePlugins(pluginManifest);
-
     const authenticatedPlugins = uniquePlugins.map((plugin) => {
       if (checkPluginAuth(plugin)) {
         return { ...plugin, authenticated: true };
@@ -123,17 +139,55 @@ const getAvailableTools = async (req, res) => {
       }
     });
 
-    const toolDefinitions = req.app.locals.availableTools;
-    const tools = authenticatedPlugins.filter(
-      (plugin) =>
-        toolDefinitions[plugin.pluginKey] !== undefined ||
-        (plugin.toolkit === true &&
-          Object.keys(toolDefinitions).some((key) => getToolkitKey(key) === plugin.pluginKey)),
-    );
+    /** Filter plugins based on availability and add MCP-specific auth config */
+    const toolsOutput = [];
+    for (const plugin of authenticatedPlugins) {
+      const isToolDefined = toolDefinitions[plugin.pluginKey] !== undefined;
+      const isToolkit =
+        plugin.toolkit === true &&
+        Object.keys(toolDefinitions).some(
+          (key) => getToolkitKey({ toolkits, toolName: key }) === plugin.pluginKey,
+        );
 
-    await cache.set(CacheKeys.TOOLS, tools);
-    res.status(200).json(tools);
+      if (!isToolDefined && !isToolkit) {
+        continue;
+      }
+
+      const toolToAdd = { ...plugin };
+
+      if (plugin.pluginKey.includes(Constants.mcp_delimiter)) {
+        const parts = plugin.pluginKey.split(Constants.mcp_delimiter);
+        const serverName = parts[parts.length - 1];
+        const serverConfig = appConfig?.mcpConfig?.[serverName];
+
+        if (serverConfig?.customUserVars) {
+          const customVarKeys = Object.keys(serverConfig.customUserVars);
+          if (customVarKeys.length === 0) {
+            toolToAdd.authConfig = [];
+            toolToAdd.authenticated = true;
+          } else {
+            toolToAdd.authConfig = Object.entries(serverConfig.customUserVars).map(
+              ([key, value]) => ({
+                authField: key,
+                label: value.title || key,
+                description: value.description || '',
+              }),
+            );
+            toolToAdd.authenticated = false;
+          }
+        }
+      }
+
+      toolsOutput.push(toolToAdd);
+    }
+
+    const finalTools = filterUniquePlugins(toolsOutput);
+    await cache.set(CacheKeys.TOOLS, finalTools);
+
+    const dedupedTools = filterUniquePlugins([...(mcpPlugins ?? []), ...finalTools]);
+    res.status(200).json(dedupedTools);
   } catch (error) {
+    logger.error('[getAvailableTools]', error);
     res.status(500).json({ message: error.message });
   }
 };

api/server/controllers/PluginController.spec.js

@@ -0,0 +1,686 @@
+const { Constants } = require('librechat-data-provider');
+const { getCachedTools, getAppConfig } = require('~/server/services/Config');
+const { getLogStores } = require('~/cache');
+
+jest.mock('@librechat/data-schemas', () => ({
+  logger: {
+    debug: jest.fn(),
+    error: jest.fn(),
+    warn: jest.fn(),
+  },
+}));
+
+jest.mock('~/server/services/Config', () => ({
+  getCachedTools: jest.fn(),
+  getAppConfig: jest.fn().mockResolvedValue({
+    filteredTools: [],
+    includedTools: [],
+  }),
+  setCachedTools: jest.fn(),
+  mergeUserTools: jest.fn(),
+}));
+
+// loadAndFormatTools mock removed - no longer used in PluginController
+
+jest.mock('~/config', () => ({
+  getMCPManager: jest.fn(() => ({
+    getAllToolFunctions: jest.fn().mockResolvedValue({}),
+    getRawConfig: jest.fn().mockReturnValue({}),
+  })),
+  getFlowStateManager: jest.fn(),
+}));
+
+jest.mock('~/app/clients/tools', () => ({
+  availableTools: [],
+  toolkits: [],
+}));
+
+jest.mock('~/cache', () => ({
+  getLogStores: jest.fn(),
+}));
+
+const { getAvailableTools, getAvailablePluginsController } = require('./PluginController');
+
+describe('PluginController', () => {
+  let mockReq, mockRes, mockCache;
+
+  beforeEach(() => {
+    jest.clearAllMocks();
+    mockReq = {
+      user: { id: 'test-user-id' },
+      config: {
+        filteredTools: [],
+        includedTools: [],
+      },
+    };
+    mockRes = { status: jest.fn().mockReturnThis(), json: jest.fn() };
+    mockCache = { get: jest.fn(), set: jest.fn() };
+    getLogStores.mockReturnValue(mockCache);
+
+    // Clear availableTools and toolkits arrays before each test
+    require('~/app/clients/tools').availableTools.length = 0;
+    require('~/app/clients/tools').toolkits.length = 0;
+
+    // Reset getCachedTools mock to ensure clean state
+    getCachedTools.mockReset();
+
+    // Reset getAppConfig mock to ensure clean state with default values
+    getAppConfig.mockReset();
+    getAppConfig.mockResolvedValue({
+      filteredTools: [],
+      includedTools: [],
+    });
+  });
+
+  describe('getAvailablePluginsController', () => {
+    it('should use filterUniquePlugins to remove duplicate plugins', async () => {
+      // Add plugins with duplicates to availableTools
+      const mockPlugins = [
+        { name: 'Plugin1', pluginKey: 'key1', description: 'First' },
+        { name: 'Plugin1', pluginKey: 'key1', description: 'First duplicate' },
+        { name: 'Plugin2', pluginKey: 'key2', description: 'Second' },
+      ];
+
+      require('~/app/clients/tools').availableTools.push(...mockPlugins);
+
+      mockCache.get.mockResolvedValue(null);
+
+      // Configure getAppConfig to return the expected config
+      getAppConfig.mockResolvedValueOnce({
+        filteredTools: [],
+        includedTools: [],
+      });
+
+      await getAvailablePluginsController(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      // The real filterUniquePlugins should have removed the duplicate
+      expect(responseData).toHaveLength(2);
+      expect(responseData[0].pluginKey).toBe('key1');
+      expect(responseData[1].pluginKey).toBe('key2');
+    });
+
+    it('should use checkPluginAuth to verify plugin authentication', async () => {
+      // checkPluginAuth returns false for plugins without authConfig
+      // so authenticated property won't be added
+      const mockPlugin = { name: 'Plugin1', pluginKey: 'key1', description: 'First' };
+
+      require('~/app/clients/tools').availableTools.push(mockPlugin);
+      mockCache.get.mockResolvedValue(null);
+
+      // Configure getAppConfig to return the expected config
+      getAppConfig.mockResolvedValueOnce({
+        filteredTools: [],
+        includedTools: [],
+      });
+
+      await getAvailablePluginsController(mockReq, mockRes);
+
+      const responseData = mockRes.json.mock.calls[0][0];
+      // The real checkPluginAuth returns false for plugins without authConfig, so authenticated property is not added
+      expect(responseData[0].authenticated).toBeUndefined();
+    });
+
+    it('should return cached plugins when available', async () => {
+      const cachedPlugins = [
+        { name: 'CachedPlugin', pluginKey: 'cached', description: 'Cached plugin' },
+      ];
+
+      mockCache.get.mockResolvedValue(cachedPlugins);
+
+      await getAvailablePluginsController(mockReq, mockRes);
+
+      // When cache is hit, we return immediately without processing
+      expect(mockRes.json).toHaveBeenCalledWith(cachedPlugins);
+    });
+
+    it('should filter plugins based on includedTools', async () => {
+      const mockPlugins = [
+        { name: 'Plugin1', pluginKey: 'key1', description: 'First' },
+        { name: 'Plugin2', pluginKey: 'key2', description: 'Second' },
+      ];
+
+      require('~/app/clients/tools').availableTools.push(...mockPlugins);
+      mockCache.get.mockResolvedValue(null);
+
+      // Configure getAppConfig to return config with includedTools
+      getAppConfig.mockResolvedValueOnce({
+        filteredTools: [],
+        includedTools: ['key1'],
+      });
+
+      await getAvailablePluginsController(mockReq, mockRes);
+
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(responseData).toHaveLength(1);
+      expect(responseData[0].pluginKey).toBe('key1');
+    });
+  });
+
+  describe('getAvailableTools', () => {
+    it('should use convertMCPToolsToPlugins for user-specific MCP tools', async () => {
+      const mockUserTools = {
+        [`tool1${Constants.mcp_delimiter}server1`]: {
+          type: 'function',
+          function: {
+            name: `tool1${Constants.mcp_delimiter}server1`,
+            description: 'Tool 1',
+            parameters: { type: 'object', properties: {} },
+          },
+        },
+      };
+
+      mockCache.get.mockResolvedValue(null);
+      getCachedTools.mockResolvedValueOnce(mockUserTools);
+      mockReq.config = {
+        mcpConfig: {
+          server1: {},
+        },
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Mock MCP manager to return empty tools initially (since getAllToolFunctions is called)
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue({}),
+        getRawConfig: jest.fn().mockReturnValue({}),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      // Mock second call to return tool definitions (includeGlobal: true)
+      getCachedTools.mockResolvedValueOnce(mockUserTools);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(responseData).toBeDefined();
+      expect(Array.isArray(responseData)).toBe(true);
+      expect(responseData.length).toBeGreaterThan(0);
+      const convertedTool = responseData.find(
+        (tool) => tool.pluginKey === `tool1${Constants.mcp_delimiter}server1`,
+      );
+      expect(convertedTool).toBeDefined();
+      // The real convertMCPToolsToPlugins extracts the name from the delimiter
+      expect(convertedTool.name).toBe('tool1');
+    });
+
+    it('should use filterUniquePlugins to deduplicate combined tools', async () => {
+      const mockUserTools = {
+        'user-tool': {
+          type: 'function',
+          function: {
+            name: 'user-tool',
+            description: 'User tool',
+            parameters: { type: 'object', properties: {} },
+          },
+        },
+      };
+
+      const mockCachedPlugins = [
+        { name: 'user-tool', pluginKey: 'user-tool', description: 'Duplicate user tool' },
+        { name: 'ManifestTool', pluginKey: 'manifest-tool', description: 'Manifest tool' },
+      ];
+
+      mockCache.get.mockResolvedValue(mockCachedPlugins);
+      getCachedTools.mockResolvedValueOnce(mockUserTools);
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Mock second call to return tool definitions
+      getCachedTools.mockResolvedValueOnce(mockUserTools);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(Array.isArray(responseData)).toBe(true);
+      // The real filterUniquePlugins should have deduplicated tools with same pluginKey
+      const userToolCount = responseData.filter((tool) => tool.pluginKey === 'user-tool').length;
+      expect(userToolCount).toBe(1);
+    });
+
+    it('should use checkPluginAuth to verify authentication status', async () => {
+      // Add a plugin to availableTools that will be checked
+      const mockPlugin = {
+        name: 'Tool1',
+        pluginKey: 'tool1',
+        description: 'Tool 1',
+        // No authConfig means checkPluginAuth returns false
+      };
+
+      require('~/app/clients/tools').availableTools.push(mockPlugin);
+
+      mockCache.get.mockResolvedValue(null);
+      // First call returns null for user tools
+      getCachedTools.mockResolvedValueOnce(null);
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Second call (with includeGlobal: true) returns the tool definitions
+      getCachedTools.mockResolvedValueOnce({
+        tool1: {
+          type: 'function',
+          function: {
+            name: 'tool1',
+            description: 'Tool 1',
+            parameters: {},
+          },
+        },
+      });
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(Array.isArray(responseData)).toBe(true);
+      const tool = responseData.find((t) => t.pluginKey === 'tool1');
+      expect(tool).toBeDefined();
+      // The real checkPluginAuth returns false for plugins without authConfig, so authenticated property is not added
+      expect(tool.authenticated).toBeUndefined();
+    });
+
+    it('should use getToolkitKey for toolkit validation', async () => {
+      const mockToolkit = {
+        name: 'Toolkit1',
+        pluginKey: 'toolkit1',
+        description: 'Toolkit 1',
+        toolkit: true,
+      };
+
+      require('~/app/clients/tools').availableTools.push(mockToolkit);
+
+      // Mock toolkits to have a mapping
+      require('~/app/clients/tools').toolkits.push({
+        name: 'Toolkit1',
+        pluginKey: 'toolkit1',
+        tools: ['toolkit1_function'],
+      });
+
+      mockCache.get.mockResolvedValue(null);
+      // First call returns null for user tools
+      getCachedTools.mockResolvedValueOnce(null);
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Second call (with includeGlobal: true) returns the tool definitions
+      getCachedTools.mockResolvedValueOnce({
+        toolkit1_function: {
+          type: 'function',
+          function: {
+            name: 'toolkit1_function',
+            description: 'Toolkit function',
+            parameters: {},
+          },
+        },
+      });
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(Array.isArray(responseData)).toBe(true);
+      const toolkit = responseData.find((t) => t.pluginKey === 'toolkit1');
+      expect(toolkit).toBeDefined();
+    });
+  });
+
+  describe('plugin.icon behavior', () => {
+    const callGetAvailableToolsWithMCPServer = async (serverConfig) => {
+      mockCache.get.mockResolvedValue(null);
+
+      const functionTools = {
+        [`test-tool${Constants.mcp_delimiter}test-server`]: {
+          type: 'function',
+          function: {
+            name: `test-tool${Constants.mcp_delimiter}test-server`,
+            description: 'A test tool',
+            parameters: { type: 'object', properties: {} },
+          },
+        },
+      };
+
+      // Mock the MCP manager to return tools and server config
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue(functionTools),
+        getRawConfig: jest.fn().mockReturnValue(serverConfig),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      // First call returns empty user tools
+      getCachedTools.mockResolvedValueOnce({});
+
+      // Mock getAppConfig to return the mcpConfig
+      mockReq.config = {
+        mcpConfig: {
+          'test-server': serverConfig,
+        },
+      };
+
+      // Second call (with includeGlobal: true) returns the tool definitions
+      getCachedTools.mockResolvedValueOnce(functionTools);
+
+      await getAvailableTools(mockReq, mockRes);
+      const responseData = mockRes.json.mock.calls[0][0];
+      return responseData.find(
+        (tool) => tool.pluginKey === `test-tool${Constants.mcp_delimiter}test-server`,
+      );
+    };
+
+    it('should set plugin.icon when iconPath is defined', async () => {
+      const serverConfig = {
+        iconPath: '/path/to/icon.png',
+      };
+      const testTool = await callGetAvailableToolsWithMCPServer(serverConfig);
+      expect(testTool.icon).toBe('/path/to/icon.png');
+    });
+
+    it('should set plugin.icon to undefined when iconPath is not defined', async () => {
+      const serverConfig = {};
+      const testTool = await callGetAvailableToolsWithMCPServer(serverConfig);
+      expect(testTool.icon).toBeUndefined();
+    });
+  });
+
+  describe('helper function integration', () => {
+    it('should properly handle MCP tools with custom user variables', async () => {
+      const appConfig = {
+        mcpConfig: {
+          'test-server': {
+            customUserVars: {
+              API_KEY: { title: 'API Key', description: 'Your API key' },
+            },
+          },
+        },
+      };
+
+      // Mock MCP tools returned by getAllToolFunctions
+      const mcpToolFunctions = {
+        [`tool1${Constants.mcp_delimiter}test-server`]: {
+          type: 'function',
+          function: {
+            name: `tool1${Constants.mcp_delimiter}test-server`,
+            description: 'Tool 1',
+            parameters: {},
+          },
+        },
+      };
+
+      // Mock the MCP manager to return tools
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue(mcpToolFunctions),
+        getRawConfig: jest.fn().mockReturnValue({
+          customUserVars: {
+            API_KEY: { title: 'API Key', description: 'Your API key' },
+          },
+        }),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      mockCache.get.mockResolvedValue(null);
+      mockReq.config = appConfig;
+
+      // First call returns user tools (empty in this case)
+      getCachedTools.mockResolvedValueOnce({});
+
+      // Second call (with includeGlobal: true) returns tool definitions including our MCP tool
+      getCachedTools.mockResolvedValueOnce(mcpToolFunctions);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(Array.isArray(responseData)).toBe(true);
+
+      // Find the MCP tool in the response
+      const mcpTool = responseData.find(
+        (tool) => tool.pluginKey === `tool1${Constants.mcp_delimiter}test-server`,
+      );
+
+      // The actual implementation adds authConfig and sets authenticated to false when customUserVars exist
+      expect(mcpTool).toBeDefined();
+      expect(mcpTool.authConfig).toEqual([
+        { authField: 'API_KEY', label: 'API Key', description: 'Your API key' },
+      ]);
+      expect(mcpTool.authenticated).toBe(false);
+    });
+
+    it('should handle error cases gracefully', async () => {
+      mockCache.get.mockRejectedValue(new Error('Cache error'));
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.json).toHaveBeenCalledWith({ message: 'Cache error' });
+    });
+  });
+
+  describe('edge cases with undefined/null values', () => {
+    it('should handle undefined cache gracefully', async () => {
+      getLogStores.mockReturnValue(undefined);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+    });
+
+    it('should handle null cachedTools and cachedUserTools', async () => {
+      mockCache.get.mockResolvedValue(null);
+      // First call returns null for user tools
+      getCachedTools.mockResolvedValueOnce(null);
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Mock MCP manager to return no tools
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue({}),
+        getRawConfig: jest.fn().mockReturnValue({}),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      // Second call (with includeGlobal: true) returns empty object instead of null
+      getCachedTools.mockResolvedValueOnce({});
+
+      await getAvailableTools(mockReq, mockRes);
+
+      // Should handle null values gracefully
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      expect(mockRes.json).toHaveBeenCalledWith([]);
+    });
+
+    it('should handle when getCachedTools returns undefined', async () => {
+      mockCache.get.mockResolvedValue(null);
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Mock getCachedTools to return undefined for both calls
+      getCachedTools.mockReset();
+      getCachedTools.mockResolvedValueOnce(undefined).mockResolvedValueOnce(undefined);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      // Should handle undefined values gracefully
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      expect(mockRes.json).toHaveBeenCalledWith([]);
+    });
+
+    it('should handle `cachedToolsArray` and `mcpPlugins` both being defined', async () => {
+      const cachedTools = [{ name: 'CachedTool', pluginKey: 'cached-tool', description: 'Cached' }];
+      // Use MCP delimiter for the user tool so convertMCPToolsToPlugins works
+      const userTools = {
+        [`user-tool${Constants.mcp_delimiter}server1`]: {
+          type: 'function',
+          function: {
+            name: `user-tool${Constants.mcp_delimiter}server1`,
+            description: 'User tool',
+            parameters: {},
+          },
+        },
+      };
+
+      mockCache.get.mockResolvedValue(cachedTools);
+      getCachedTools.mockResolvedValueOnce(userTools);
+      mockReq.config = {
+        mcpConfig: {
+          server1: {},
+        },
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Mock MCP manager to return empty tools initially
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue({}),
+        getRawConfig: jest.fn().mockReturnValue({}),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      // The controller expects a second call to getCachedTools
+      getCachedTools.mockResolvedValueOnce({
+        'cached-tool': { type: 'function', function: { name: 'cached-tool' } },
+        [`user-tool${Constants.mcp_delimiter}server1`]:
+          userTools[`user-tool${Constants.mcp_delimiter}server1`],
+      });
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      // Should have both cached and user tools
+      expect(responseData.length).toBeGreaterThanOrEqual(2);
+    });
+
+    it('should handle empty toolDefinitions object', async () => {
+      mockCache.get.mockResolvedValue(null);
+      // Reset getCachedTools to ensure clean state
+      getCachedTools.mockReset();
+      getCachedTools.mockResolvedValue({});
+      mockReq.config = {}; // No mcpConfig at all
+
+      // Ensure no plugins are available
+      require('~/app/clients/tools').availableTools.length = 0;
+
+      // Reset MCP manager to default state
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue({}),
+        getRawConfig: jest.fn().mockReturnValue({}),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      await getAvailableTools(mockReq, mockRes);
+
+      // With empty tool definitions, no tools should be in the final output
+      expect(mockRes.json).toHaveBeenCalledWith([]);
+    });
+
+    it('should handle MCP tools without customUserVars', async () => {
+      const appConfig = {
+        mcpConfig: {
+          'test-server': {
+            // No customUserVars defined
+          },
+        },
+      };
+
+      const mockUserTools = {
+        [`tool1${Constants.mcp_delimiter}test-server`]: {
+          type: 'function',
+          function: {
+            name: `tool1${Constants.mcp_delimiter}test-server`,
+            description: 'Tool 1',
+            parameters: { type: 'object', properties: {} },
+          },
+        },
+      };
+
+      // Mock the MCP manager to return the tools
+      const mockMCPManager = {
+        getAllToolFunctions: jest.fn().mockResolvedValue(mockUserTools),
+        getRawConfig: jest.fn().mockReturnValue({
+          // No customUserVars defined
+        }),
+      };
+      require('~/config').getMCPManager.mockReturnValue(mockMCPManager);
+
+      mockCache.get.mockResolvedValue(null);
+      mockReq.config = appConfig;
+      // First call returns empty user tools
+      getCachedTools.mockResolvedValueOnce({});
+
+      // Second call (with includeGlobal: true) returns the tool definitions
+      getCachedTools.mockResolvedValueOnce(mockUserTools);
+
+      // Ensure no plugins in availableTools for clean test
+      require('~/app/clients/tools').availableTools.length = 0;
+
+      await getAvailableTools(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      const responseData = mockRes.json.mock.calls[0][0];
+      expect(Array.isArray(responseData)).toBe(true);
+      expect(responseData.length).toBeGreaterThan(0);
+
+      const mcpTool = responseData.find(
+        (tool) => tool.pluginKey === `tool1${Constants.mcp_delimiter}test-server`,
+      );
+
+      expect(mcpTool).toBeDefined();
+      expect(mcpTool.authenticated).toBe(true);
+      // The actual implementation sets authConfig to empty array when no customUserVars
+      expect(mcpTool.authConfig).toEqual([]);
+    });
+
+    it('should handle undefined filteredTools and includedTools', async () => {
+      mockReq.config = {};
+      mockCache.get.mockResolvedValue(null);
+
+      // Configure getAppConfig to return config with undefined properties
+      // The controller will use default values [] for filteredTools and includedTools
+      getAppConfig.mockResolvedValueOnce({});
+
+      await getAvailablePluginsController(mockReq, mockRes);
+
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+      expect(mockRes.json).toHaveBeenCalledWith([]);
+    });
+
+    it('should handle toolkit with undefined toolDefinitions keys', async () => {
+      const mockToolkit = {
+        name: 'Toolkit1',
+        pluginKey: 'toolkit1',
+        description: 'Toolkit 1',
+        toolkit: true,
+      };
+
+      // No need to mock app.locals anymore as it's not used
+
+      // Add the toolkit to availableTools
+      require('~/app/clients/tools').availableTools.push(mockToolkit);
+
+      mockCache.get.mockResolvedValue(null);
+      // First call returns empty object
+      getCachedTools.mockResolvedValueOnce({});
+      mockReq.config = {
+        mcpConfig: null,
+        paths: { structuredTools: '/mock/path' },
+      };
+
+      // Second call (with includeGlobal: true) returns empty object to avoid null reference error
+      getCachedTools.mockResolvedValueOnce({});
+
+      await getAvailableTools(mockReq, mockRes);
+
+      // Should handle null toolDefinitions gracefully
+      expect(mockRes.status).toHaveBeenCalledWith(200);
+    });
+  });
+});

api/server/controllers/TwoFactorController.js

@@ -1,3 +1,4 @@
+const { encryptV3 } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
 const {
   verifyTOTP,
@@ -7,7 +8,6 @@ const {
   generateBackupCodes,
 } = require('~/server/services/twoFactorService');
 const { getUserById, updateUser } = require('~/models');
-const { encryptV3 } = require('~/server/utils/crypto');
 
 const safeAppTitle = (process.env.APP_TITLE || 'LibreChat').replace(/\s+/g, '');
 
@@ -47,7 +47,7 @@ const verify2FA = async (req, res) => {
   try {
     const userId = req.user.id;
     const { token, backupCode } = req.body;
-    const user = await getUserById(userId);
+    const user = await getUserById(userId, '_id totpSecret backupCodes');
 
     if (!user || !user.totpSecret) {
       return res.status(400).json({ message: '2FA not initiated' });
@@ -79,7 +79,7 @@ const confirm2FA = async (req, res) => {
   try {
     const userId = req.user.id;
     const { token } = req.body;
-    const user = await getUserById(userId);
+    const user = await getUserById(userId, '_id totpSecret');
 
     if (!user || !user.totpSecret) {
       return res.status(400).json({ message: '2FA not initiated' });
@@ -99,10 +99,36 @@ const confirm2FA = async (req, res) => {
 
 /**
  * Disable 2FA by clearing the stored secret and backup codes.
+ * Requires verification with either TOTP token or backup code if 2FA is fully enabled.
  */
 const disable2FA = async (req, res) => {
   try {
     const userId = req.user.id;
+    const { token, backupCode } = req.body;
+    const user = await getUserById(userId, '_id totpSecret backupCodes');
+
+    if (!user || !user.totpSecret) {
+      return res.status(400).json({ message: '2FA is not setup for this user' });
+    }
+
+    if (user.twoFactorEnabled) {
+      const secret = await getTOTPSecret(user.totpSecret);
+      let isVerified = false;
+
+      if (token) {
+        isVerified = await verifyTOTP(secret, token);
+      } else if (backupCode) {
+        isVerified = await verifyBackupCode({ user, backupCode });
+      } else {
+        return res
+          .status(400)
+          .json({ message: 'Either token or backup code is required to disable 2FA' });
+      }
+
+      if (!isVerified) {
+        return res.status(401).json({ message: 'Invalid token or backup code' });
+      }
+    }
     await updateUser(userId, { totpSecret: null, backupCodes: [], twoFactorEnabled: false });
     return res.status(200).json();
   } catch (err) {

api/server/controllers/UserController.js

@@ -1,10 +1,10 @@
+const { logger } = require('@librechat/data-schemas');
 const {
-  Tools,
-  FileSources,
   webSearchKeys,
   extractWebSearchEnvVars,
-} = require('librechat-data-provider');
-const { logger } = require('@librechat/data-schemas');
+  normalizeHttpError,
+  MCPTokenStorage,
+} = require('@librechat/api');
 const {
   getFiles,
   updateUser,
@@ -19,16 +19,32 @@ const { updateUserPluginAuth, deleteUserPluginAuth } = require('~/server/service
 const { updateUserPluginsService, deleteUserKey } = require('~/server/services/UserService');
 const { verifyEmail, resendVerificationEmail } = require('~/server/services/AuthService');
 const { needsRefresh, getNewS3URL } = require('~/server/services/Files/S3/crud');
+const { Tools, Constants, FileSources } = require('librechat-data-provider');
 const { processDeleteRequest } = require('~/server/services/Files/process');
-const { Transaction, Balance, User } = require('~/db/models');
+const { Transaction, Balance, User, Token } = require('~/db/models');
+const { getAppConfig } = require('~/server/services/Config');
 const { deleteToolCalls } = require('~/models/ToolCall');
 const { deleteAllSharedLinks } = require('~/models');
+const { getMCPManager } = require('~/config');
+const { MCPOAuthHandler } = require('@librechat/api');
+const { getFlowStateManager } = require('~/config');
+const { CacheKeys } = require('librechat-data-provider');
+const { getLogStores } = require('~/cache');
+const { clearMCPServerTools } = require('~/server/services/Config/mcpToolsCache');
+const { findToken } = require('~/models');
 
 const getUserController = async (req, res) => {
-  /** @type {MongoUser} */
+  const appConfig = await getAppConfig({ role: req.user?.role });
+  /** @type {IUser} */
   const userData = req.user.toObject != null ? req.user.toObject() : { ...req.user };
+  /**
+   * These fields should not exist due to secure field selection, but deletion
+   * is done in case of alternate database incompatibility with Mongo API
+   * */
+  delete userData.password;
   delete userData.totpSecret;
-  if (req.app.locals.fileStrategy === FileSources.s3 && userData.avatar) {
+  delete userData.backupCodes;
+  if (appConfig.fileStrategy === FileSources.s3 && userData.avatar) {
     const avatarNeedsRefresh = needsRefresh(userData.avatar, 3600);
     if (!avatarNeedsRefresh) {
       return res.status(200).send(userData);
@@ -84,6 +100,7 @@ const deleteUserFiles = async (req) => {
 };
 
 const updateUserPluginsController = async (req, res) => {
+  const appConfig = await getAppConfig({ role: req.user?.role });
   const { user } = req;
   const { pluginKey, action, auth, isEntityTool } = req.body;
   try {
@@ -92,8 +109,8 @@ const updateUserPluginsController = async (req, res) => {
 
       if (userPluginsService instanceof Error) {
         logger.error('[userPluginsService]', userPluginsService);
-        const { status, message } = userPluginsService;
-        res.status(status).send({ message });
+        const { status, message } = normalizeHttpError(userPluginsService);
+        return res.status(status).send({ message });
       }
     }
 
@@ -102,10 +119,22 @@ const updateUserPluginsController = async (req, res) => {
     }
 
     let keys = Object.keys(auth);
-    if (keys.length === 0 && pluginKey !== Tools.web_search) {
+    const values = Object.values(auth); // Used in 'install' block
+
+    const isMCPTool = pluginKey.startsWith('mcp_') || pluginKey.includes(Constants.mcp_delimiter);
+
+    // Early exit condition:
+    // If keys are empty (meaning auth: {} was likely sent for uninstall, or auth was empty for install)
+    // AND it's not web_search (which has special key handling to populate `keys` for uninstall)
+    // AND it's NOT (an uninstall action FOR an MCP tool - we need to proceed for this case to clear all its auth)
+    // THEN return.
+    if (
+      keys.length === 0 &&
+      pluginKey !== Tools.web_search &&
+      !(action === 'uninstall' && isMCPTool)
+    ) {
       return res.status(200).send();
     }
-    const values = Object.values(auth);
 
     /** @type {number} */
     let status = 200;
@@ -116,7 +145,7 @@ const updateUserPluginsController = async (req, res) => {
 
     if (pluginKey === Tools.web_search) {
       /** @type  {TCustomConfig['webSearch']} */
-      const webSearchConfig = req.app.locals?.webSearch;
+      const webSearchConfig = appConfig?.webSearch;
       keys = extractWebSearchEnvVars({
         keys: action === 'install' ? keys : webSearchKeys,
         config: webSearchConfig,
@@ -128,24 +157,73 @@ const updateUserPluginsController = async (req, res) => {
         authService = await updateUserPluginAuth(user.id, keys[i], pluginKey, values[i]);
         if (authService instanceof Error) {
           logger.error('[authService]', authService);
-          ({ status, message } = authService);
+          ({ status, message } = normalizeHttpError(authService));
         }
       }
     } else if (action === 'uninstall') {
-      for (let i = 0; i < keys.length; i++) {
-        authService = await deleteUserPluginAuth(user.id, keys[i]);
+      // const isMCPTool was defined earlier
+      if (isMCPTool && keys.length === 0) {
+        // This handles the case where auth: {} is sent for an MCP tool uninstall.
+        // It means "delete all credentials associated with this MCP pluginKey".
+        authService = await deleteUserPluginAuth(user.id, null, true, pluginKey);
         if (authService instanceof Error) {
-          logger.error('[authService]', authService);
-          ({ status, message } = authService);
+          logger.error(
+            `[authService] Error deleting all auth for MCP tool ${pluginKey}:`,
+            authService,
+          );
+          ({ status, message } = normalizeHttpError(authService));
+        }
+        try {
+          // if the MCP server uses OAuth, perform a full cleanup and token revocation
+          await maybeUninstallOAuthMCP(user.id, pluginKey, appConfig);
+        } catch (error) {
+          logger.error(
+            `[updateUserPluginsController] Error uninstalling OAuth MCP for ${pluginKey}:`,
+            error,
+          );
+        }
+      } else {
+        // This handles:
+        // 1. Web_search uninstall (keys will be populated with all webSearchKeys if auth was {}).
+        // 2. Other tools uninstall (if keys were provided).
+        // 3. MCP tool uninstall if specific keys were provided in `auth` (not current frontend behavior).
+        // If keys is empty for non-MCP tools (and not web_search), this loop won't run, and nothing is deleted.
+        for (let i = 0; i < keys.length; i++) {
+          authService = await deleteUserPluginAuth(user.id, keys[i]); // Deletes by authField name
+          if (authService instanceof Error) {
+            logger.error('[authService] Error deleting specific auth key:', authService);
+            ({ status, message } = normalizeHttpError(authService));
+          }
         }
       }
     }
 
     if (status === 200) {
+      // If auth was updated successfully, disconnect MCP sessions as they might use these credentials
+      if (pluginKey.startsWith(Constants.mcp_prefix)) {
+        try {
+          const mcpManager = getMCPManager(user.id);
+          if (mcpManager) {
+            // Extract server name from pluginKey (format: "mcp_<serverName>")
+            const serverName = pluginKey.replace(Constants.mcp_prefix, '');
+            logger.info(
+              `[updateUserPluginsController] Attempting disconnect of MCP server "${serverName}" for user ${user.id} after plugin auth update.`,
+            );
+            await mcpManager.disconnectUserConnection(user.id, serverName);
+          }
+        } catch (disconnectError) {
+          logger.error(
+            `[updateUserPluginsController] Error disconnecting MCP connection for user ${user.id} after plugin auth update:`,
+            disconnectError,
+          );
+          // Do not fail the request for this, but log it.
+        }
+      }
       return res.status(status).send();
     }
 
-    res.status(status).send({ message });
+    const normalized = normalizeHttpError({ status, message });
+    return res.status(normalized.status).send({ message: normalized.message });
   } catch (err) {
     logger.error('[updateUserPluginsController]', err);
     return res.status(500).json({ message: 'Something went wrong.' });
@@ -211,6 +289,97 @@ const resendVerificationController = async (req, res) => {
   }
 };
 
+/**
+ * OAuth MCP specific uninstall logic
+ */
+const maybeUninstallOAuthMCP = async (userId, pluginKey, appConfig) => {
+  if (!pluginKey.startsWith(Constants.mcp_prefix)) {
+    // this is not an MCP server, so nothing to do here
+    return;
+  }
+
+  const serverName = pluginKey.replace(Constants.mcp_prefix, '');
+  const mcpManager = getMCPManager(userId);
+  const serverConfig = mcpManager.getRawConfig(serverName) ?? appConfig?.mcpServers?.[serverName];
+
+  if (!mcpManager.getOAuthServers().has(serverName)) {
+    // this server does not use OAuth, so nothing to do here as well
+    return;
+  }
+
+  // 1. get client info used for revocation (client id, secret)
+  const clientTokenData = await MCPTokenStorage.getClientInfoAndMetadata({
+    userId,
+    serverName,
+    findToken,
+  });
+  if (clientTokenData == null) {
+    return;
+  }
+  const { clientInfo, clientMetadata } = clientTokenData;
+
+  // 2. get decrypted tokens before deletion
+  const tokens = await MCPTokenStorage.getTokens({
+    userId,
+    serverName,
+    findToken,
+  });
+
+  // 3. revoke OAuth tokens at the provider
+  const revocationEndpoint =
+    serverConfig.oauth?.revocation_endpoint ?? clientMetadata.revocation_endpoint;
+  const revocationEndpointAuthMethodsSupported =
+    serverConfig.oauth?.revocation_endpoint_auth_methods_supported ??
+    clientMetadata.revocation_endpoint_auth_methods_supported;
+
+  if (tokens?.access_token) {
+    try {
+      await MCPOAuthHandler.revokeOAuthToken(serverName, tokens.access_token, 'access', {
+        serverUrl: serverConfig.url,
+        clientId: clientInfo.client_id,
+        clientSecret: clientInfo.client_secret ?? '',
+        revocationEndpoint,
+        revocationEndpointAuthMethodsSupported,
+      });
+    } catch (error) {
+      logger.error(`Error revoking OAuth access token for ${serverName}:`, error);
+    }
+  }
+
+  if (tokens?.refresh_token) {
+    try {
+      await MCPOAuthHandler.revokeOAuthToken(serverName, tokens.refresh_token, 'refresh', {
+        serverUrl: serverConfig.url,
+        clientId: clientInfo.client_id,
+        clientSecret: clientInfo.client_secret ?? '',
+        revocationEndpoint,
+        revocationEndpointAuthMethodsSupported,
+      });
+    } catch (error) {
+      logger.error(`Error revoking OAuth refresh token for ${serverName}:`, error);
+    }
+  }
+
+  // 4. delete tokens from the DB after revocation attempts
+  await MCPTokenStorage.deleteUserTokens({
+    userId,
+    serverName,
+    deleteToken: async (filter) => {
+      await Token.deleteOne(filter);
+    },
+  });
+
+  // 5. clear the flow state for the OAuth tokens
+  const flowsCache = getLogStores(CacheKeys.FLOWS);
+  const flowManager = getFlowStateManager(flowsCache);
+  const flowId = MCPOAuthHandler.generateFlowId(userId, serverName);
+  await flowManager.deleteFlow(flowId, 'mcp_get_tokens');
+  await flowManager.deleteFlow(flowId, 'mcp_oauth');
+
+  // 6. clear the tools cache for the server
+  await clearMCPServerTools({ userId, serverName });
+};
+
 module.exports = {
   getUserController,
   getTermsStatusController,