chore: Remove unused permission-related UI localization keys

- Added `useInfiniteScroll` hook to manage infinite scrolling behavior in agent grids.
- Integrated infinite scroll functionality into `AgentGrid` and `VirtualizedAgentGrid` components.
- Updated `AgentMarketplace` to pass the scroll container to the agent grid components.
- Refactored loading indicators to show a spinner instead of a "Load More" button.
- Created `VirtualizedAgentGrid` component for optimized rendering of agent cards using virtualization.
- Added performance tests for `VirtualizedAgentGrid` to ensure efficient handling of large datasets.
- Updated translations to include new messages for end-of-results scenarios.

.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
@@ -58,7 +72,7 @@ DEBUG_CONSOLE=false
 #                     Endpoints                     #
 #===================================================#
 
-# ENDPOINTS=openAI,assistants,azureOpenAI,google,gptPlugins,anthropic
+# ENDPOINTS=openAI,assistants,azureOpenAI,google,anthropic
 
 PROXY=
 
@@ -142,10 +156,10 @@ 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 +363,11 @@ 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 +456,8 @@ 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 +474,26 @@ 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.
@@ -590,6 +626,10 @@ 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                         #
 #===================================================#
@@ -607,11 +647,40 @@ 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
+
+# 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., STATIC_CONFIG,ROLES,MESSAGES)
+# FORCED_IN_MEMORY_CACHE_NAMESPACES=STATIC_CONFIG,ROLES
 
 #==================================================#
 #                      Others                      #
@@ -672,4 +741,4 @@ 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

.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-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/i18n-unused-keys.yml

@@ -6,6 +6,7 @@ on:
       - "client/src/**"
       - "api/**"
       - "packages/data-provider/src/**"
+      - "packages/client/**"
 
 jobs:
   detect-unused-i18n-keys:
@@ -23,7 +24,7 @@ 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")
 
           # Check if translation file exists
           if [[ ! -f "$I18N_FILE" ]]; then

.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" >> $GITHUB_ENV
@@ -97,7 +184,8 @@ jobs:
             chmod -R 755 client
             cd client
             UNUSED=$(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" >> $GITHUB_ENV
@@ -113,7 +201,8 @@ jobs:
             chmod -R 755 api
             cd api
             UNUSED=$(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

@@ -125,3 +125,12 @@ helm/**/.values.yaml
 
 # SAML Idp cert
 *.cert
+
+# AI Assistants
+/.claude/
+/.cursor/
+/.copilot/
+/.aider/
+/.openai/
+/.tabnine/
+/.codeium

Dockerfile

@@ -1,4 +1,4 @@
-# v0.7.8
+# v0.8.0-rc1
 
 # Base node image
 FROM node:20-alpine AS node

Dockerfile.multi

@@ -1,5 +1,5 @@
 # Dockerfile.multi
-# v0.7.8
+# v0.8.0-rc1
 
 # 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,
@@ -66,10 +66,9 @@
 - 🔦 **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
+    - 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

api/app/clients/BaseClient.js

@@ -13,7 +13,6 @@ 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');
@@ -109,12 +108,15 @@ 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 {number} promptTokens
    * @param {number} completionTokens
    * @returns {Promise<void>}
    */
-  async recordTokenUsage({ promptTokens, completionTokens }) {
+  async recordTokenUsage({ model, promptTokens, completionTokens }) {
     logger.debug('[BaseClient] `recordTokenUsage` not implemented.', {
+      model,
       promptTokens,
       completionTokens,
     });
@@ -198,6 +200,10 @@ class BaseClient {
       this.currentMessages[this.currentMessages.length - 1].messageId = head;
     }
 
+    if (opts.isRegenerate && responseMessageId.endsWith('_')) {
+      responseMessageId = crypto.randomUUID();
+    }
+
     this.responseMessageId = responseMessageId;
 
     return {
@@ -572,7 +578,7 @@ class BaseClient {
       });
     }
 
-    const { generation = '' } = opts;
+    const { editedContent } = opts;
 
     // It's not necessary to push to currentMessages
     // depending on subclass implementation of handling messages
@@ -587,11 +593,21 @@ 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 {
@@ -672,16 +688,32 @@ 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 +744,13 @@ class BaseClient {
       } else {
         responseMessage.tokenCount = this.getTokenCountForResponse(responseMessage);
         completionTokens = responseMessage.tokenCount;
+        await this.recordTokenUsage({
+          usage,
+          promptTokens,
+          completionTokens,
+          model: responseMessage.model,
+        });
       }
-
-      await this.recordTokenUsage({ promptTokens, completionTokens, usage });
     }
 
     if (userMessagePromise) {
@@ -1095,6 +1131,50 @@ 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 +1,7 @@
 const { google } = require('googleapis');
-const { Tokenizer } = 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');
@@ -12,13 +12,13 @@ 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');
@@ -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

@@ -5,6 +5,7 @@ const {
   isEnabled,
   Tokenizer,
   createFetch,
+  resolveHeaders,
   constructAzureURL,
   genAzureChatCompletion,
   createStreamEventHandlers,
@@ -15,7 +16,6 @@ const {
   ContentTypes,
   parseTextParts,
   EModelEndpoint,
-  resolveHeaders,
   KnownEndpoints,
   openAISettings,
   ImageDetailCost,
@@ -37,7 +37,6 @@ 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 { tokenSplit } = require('./document');
@@ -47,12 +46,6 @@ 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 +372,12 @@ 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;

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/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/prompts/createContextHandlers.js

@@ -1,6 +1,7 @@
 const axios = require('axios');
-const { isEnabled } = require('~/server/utils');
-const { logger } = require('~/config');
+const { isEnabled } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
+const { generateShortLivedToken } = require('~/server/services/AuthService');
 
 const footer = `Use the context as your learned knowledge to better answer the user.
 
@@ -18,7 +19,7 @@ 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) => {

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/BaseClient.test.js

@@ -422,6 +422,46 @@ 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 = {

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/structured/DALLE3.js

@@ -3,8 +3,8 @@ 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 { FileContext, ContentTypes } = require('librechat-data-provider');
 const { getImageBasename } = require('~/server/services/Files/images');
 const extractBaseURL = require('~/utils/extractBaseURL');
@@ -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

@@ -3,10 +3,10 @@ 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 { ContentTypes, EImageOutputType } = require('librechat-data-provider');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
 const { extractBaseURL } = require('~/utils');
@@ -107,6 +107,12 @@ 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)
  * @param {Object} fields - Configuration fields
@@ -183,7 +189,10 @@ 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 +210,18 @@ 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 +245,10 @@ 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) {
@@ -317,7 +338,10 @@ 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 +433,17 @@ 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 +453,19 @@ 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,
@@ -467,6 +511,10 @@ 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);
+        }
       }
     },
     {

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

@@ -0,0 +1,94 @@
+const DALLE3 = require('../DALLE3');
+const { ProxyAgent } = require('undici');
+
+const processFileURL = jest.fn();
+
+jest.mock('~/server/services/Files/images', () => ({
+  getImageBasename: jest.fn().mockImplementation((url) => {
+    const parts = url.split('/');
+    const lastPart = parts.pop();
+    const imageExtensionRegex = /\.(jpg|jpeg|png|gif|bmp|tiff|svg)$/i;
+    if (imageExtensionRegex.test(lastPart)) {
+      return lastPart;
+    }
+    return '';
+  }),
+}));
+
+jest.mock('fs', () => {
+  return {
+    existsSync: jest.fn(),
+    mkdirSync: jest.fn(),
+    promises: {
+      writeFile: jest.fn(),
+      readFile: jest.fn(),
+      unlink: jest.fn(),
+    },
+  };
+});
+
+jest.mock('path', () => {
+  return {
+    resolve: jest.fn(),
+    join: jest.fn(),
+    relative: jest.fn(),
+    extname: jest.fn().mockImplementation((filename) => {
+      return filename.slice(filename.lastIndexOf('.'));
+    }),
+  };
+});
+
+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/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 { Tools, EToolResources } = require('librechat-data-provider');
+const { filterFilesByAgentAccess } = require('~/server/services/Files/permissions');
+const { generateShortLivedToken } = require('~/server/services/AuthService');
 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.`;
 
@@ -59,7 +79,7 @@ const createFileSearchTool = async ({ req, files, entity_id }) => {
       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 +125,13 @@ 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 +141,37 @@ const createFileSearchTool = async ({ req, files, entity_id }) => {
 
       const formattedString = formattedResults
         .map(
-          (result) =>
-            `File: ${result.filename}\nRelevance: ${1.0 - result.distance.toFixed(4)}\nContent: ${
+          (result, index) =>
+            `File: ${result.filename}\nAnchor: \\ue202turn0file${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 } }];
     },
     {
       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.
+
+**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 +1,9 @@
-const { mcpToolPattern } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
 const { SerpAPI } = require('@langchain/community/tools/serpapi');
 const { Calculator } = require('@langchain/community/tools/calculator');
+const { mcpToolPattern, loadWebSearchAuth } = require('@librechat/api');
 const { EnvVar, createCodeExecutionTool, createSearchTool } = require('@librechat/agents');
-const {
-  Tools,
-  EToolResources,
-  loadWebSearchAuth,
-  replaceSpecialVars,
-} = require('librechat-data-provider');
+const { Tools, EToolResources, replaceSpecialVars } = require('librechat-data-provider');
 const {
   availableTools,
   manifestToolMap,
@@ -235,7 +230,7 @@ const loadTools = async ({
 
   /** @type {Record<string, string>} */
   const toolContextMap = {};
-  const appTools = (await getCachedTools({ includeGlobal: true })) ?? {};
+  const cachedTools = (await getCachedTools({ userId: user, includeGlobal: true })) ?? {};
 
   for (const tool of tools) {
     if (tool === Tools.execute_code) {
@@ -245,7 +240,13 @@ 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;
         }
@@ -260,7 +261,10 @@ 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;
         }
@@ -294,7 +298,7 @@ Current Date & Time: ${replaceSpecialVars({ text: '{{iso_datetime}}' })}
         });
       };
       continue;
-    } else if (tool && appTools[tool] && mcpToolPattern.test(tool)) {
+    } else if (tool && cachedTools && mcpToolPattern.test(tool)) {
       requestedTools[tool] = async () =>
         createMCPTool({
           req: options.req,

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 +1,62 @@
+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'),
+
+  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 +1,157 @@
+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.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('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, STATIC_CONFIG ,MESSAGES ';
+
+      const { cacheConfig } = require('./cacheConfig');
+      expect(cacheConfig.FORCED_IN_MEMORY_CACHE_NAMESPACES).toEqual([
+        'ROLES',
+        'STATIC_CONFIG',
+        '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

@@ -1,113 +1,53 @@
+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 mcpTools = isRedisEnabled
-  ? new Keyv({ store: keyvRedis })
-  : new Keyv({ namespace: CacheKeys.MCP_TOOLS });
-
-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.MCP_TOOLS]: mcpTools,
-  [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.MCP_TOOLS]: standardCache(CacheKeys.MCP_TOOLS),
+  [CacheKeys.CONFIG_STORE]: standardCache(CacheKeys.CONFIG_STORE),
+  [CacheKeys.STATIC_CONFIG]: standardCache(CacheKeys.STATIC_CONFIG),
+  [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,
 };
 
 /**
@@ -116,7 +56,10 @@ 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
   );
 }
 
@@ -152,18 +95,18 @@ 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;
       }
@@ -172,7 +115,7 @@ 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}`,
       );
@@ -213,7 +156,7 @@ async function clearAllExpiredFromCache() {
   }
 }
 
-if (!isRedisEnabled && !isEnabled(CI)) {
+if (!cacheConfig.USE_REDIS && !cacheConfig.CI) {
   /** @type {Set<NodeJS.Timeout>} */
   const cleanupIntervals = new Set();
 
@@ -224,7 +167,7 @@ if (!isRedisEnabled && !isEnabled(CI)) {
 
   cleanupIntervals.add(cleanup);
 
-  if (debugMemoryCache) {
+  if (cacheConfig.DEBUG_MEMORY_CACHE) {
     const monitor = setInterval(() => {
       const ttlStores = getTTLStores();
       const memory = process.memoryUsage();
@@ -245,13 +188,13 @@ 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 +1,204 @@
+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 true;
+      }
+      return false;
+    },
+    enableOfflineQueue: cacheConfig.REDIS_ENABLE_OFFLINE_QUEUE,
+    connectTimeout: cacheConfig.REDIS_CONNECT_TIMEOUT,
+    maxRetriesPerRequest: 3,
+  };
+
+  ioredisClient =
+    urls.length === 1
+      ? new IoRedis(cacheConfig.REDIS_URI, redisOptions)
+      : new IoRedis.Cluster(cacheConfig.REDIS_URI, {
+          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
+      ? createClient({ url: cacheConfig.REDIS_URI, ...redisOptions })
+      : createCluster({
+          rootNodes: cacheConfig.REDIS_URI.split(',').map((url) => ({ url })),
+          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/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/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

@@ -1,15 +1,16 @@
 const mongoose = require('mongoose');
 const crypto = require('node:crypto');
 const { logger } = require('@librechat/data-schemas');
-const { SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
+const { ResourceType, SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
 const { GLOBAL_PROJECT_NAME, EPHEMERAL_AGENT_ID, mcp_delimiter } =
   require('librechat-data-provider').Constants;
 const {
-  getProjectByName,
-  addAgentIdsToProject,
-  removeAgentIdsFromProject,
   removeAgentFromAllProjects,
+  removeAgentIdsFromProject,
+  addAgentIdsToProject,
+  getProjectByName,
 } = require('./Project');
+const { removeAllPermissions } = require('~/server/services/PermissionService');
 const { getCachedTools } = require('~/server/services/Config');
 const { getActions } = require('./Action');
 const { Agent } = require('~/db/models');
@@ -32,7 +33,9 @@ const createAgent = async (agentData) => {
         updatedAt: timestamp,
       },
     ],
+    category: agentData.category || 'general',
   };
+
   return (await Agent.create(initialAgentData)).toObject();
 };
 
@@ -59,7 +62,7 @@ const getAgent = async (searchParameter) => await Agent.findOne(searchParameter)
 const loadEphemeralAgent = async ({ req, agent_id, endpoint, model_parameters: _m }) => {
   const { model, ...model_parameters } = _m;
   /** @type {Record<string, FunctionTool>} */
-  const availableTools = await getCachedTools({ includeGlobal: true });
+  const availableTools = await getCachedTools({ userId: req.user.id, includeGlobal: true });
   /** @type {TEphemeralAgent | null} */
   const ephemeralAgent = req.body.ephemeralAgent;
   const mcpServers = new Set(ephemeralAgent?.mcp);
@@ -88,7 +91,7 @@ const loadEphemeralAgent = async ({ req, agent_id, endpoint, model_parameters: _
   }
 
   const instructions = req.body.promptPrefix;
-  return {
+  const result = {
     id: agent_id,
     instructions,
     provider: endpoint,
@@ -96,6 +99,11 @@ const loadEphemeralAgent = async ({ req, agent_id, endpoint, model_parameters: _
     model,
     tools,
   };
+
+  if (ephemeralAgent?.artifacts != null && ephemeralAgent.artifacts) {
+    result.artifacts = ephemeralAgent.artifacts;
+  }
+  return result;
 };
 
 /**
@@ -173,54 +181,116 @@ 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;
     }
   }
 
@@ -443,6 +513,10 @@ 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;
 };
@@ -466,11 +540,7 @@ const getListAgentsByAccess = async ({
   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 };
-
-  if (accessibleIds.length > 0) {
-    baseQuery._id = { $in: accessibleIds };
-  }
+  const baseQuery = { ...otherParams, _id: { $in: accessibleIds } };
 
   // Add cursor condition
   if (after) {
@@ -481,7 +551,7 @@ const getListAgentsByAccess = async ({
       const cursorCondition = {
         $or: [
           { updatedAt: { $lt: new Date(updatedAt) } },
-          { updatedAt: new Date(updatedAt), _id: { $gt: mongoose.Types.ObjectId(_id) } },
+          { updatedAt: new Date(updatedAt), _id: { $gt: new mongoose.Types.ObjectId(_id) } },
         ],
       };
 
@@ -509,6 +579,9 @@ const getListAgentsByAccess = async ({
     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
@@ -577,6 +650,7 @@ const getListAgents = async (searchParameter) => {
       description: 1,
       // @deprecated - isCollaborative replaced by ACL permissions
       isCollaborative: 1,
+      category: 1,
     }).lean()
   ).map((agent) => {
     if (agent.author?.toString() !== author) {
@@ -742,6 +816,14 @@ 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
@@ -762,4 +844,5 @@ module.exports = {
   getListAgentsByAccess,
   removeAgentResourceFiles,
   generateActionMetadataHash,
+  countPromotedAgents,
 };

api/models/Agent.spec.js

@@ -14,6 +14,7 @@ const mongoose = require('mongoose');
 const { v4: uuidv4 } = require('uuid');
 const { agentSchema } = require('@librechat/data-schemas');
 const { MongoMemoryServer } = require('mongodb-memory-server');
+const { AccessRoleIds, ResourceType, PrincipalType } = require('librechat-data-provider');
 const {
   getAgent,
   loadAgent,
@@ -21,13 +22,16 @@ const {
   updateAgent,
   deleteAgent,
   getListAgents,
+  getListAgentsByAccess,
+  revertAgentVersion,
   updateAgentProjects,
   addAgentResourceFile,
   removeAgentResourceFiles,
   generateActionMetadataHash,
-  revertAgentVersion,
 } = require('./Agent');
+const permissionService = require('~/server/services/PermissionService');
 const { getCachedTools } = require('~/server/services/Config');
+const { AclEntry } = require('~/db/models');
 
 /**
  * @type {import('mongoose').Model<import('@librechat/data-schemas').IAgent>}
@@ -407,12 +411,26 @@ describe('models/Agent', () => {
 
   describe('Agent CRUD Operations', () => {
     let mongoServer;
+    let AccessRole;
 
     beforeAll(async () => {
       mongoServer = await MongoMemoryServer.create();
       const mongoUri = mongoServer.getUri();
       Agent = mongoose.models.Agent || mongoose.model('Agent', agentSchema);
       await mongoose.connect(mongoUri);
+
+      // Initialize models
+      const dbModels = require('~/db/models');
+      AccessRole = dbModels.AccessRole;
+
+      // Create necessary access roles for agents
+      await AccessRole.create({
+        accessRoleId: AccessRoleIds.AGENT_OWNER,
+        name: 'Owner',
+        description: 'Full control over agents',
+        resourceType: ResourceType.AGENT,
+        permBits: 15, // VIEW | EDIT | DELETE | SHARE
+      });
     }, 20000);
 
     afterAll(async () => {
@@ -468,6 +486,51 @@ describe('models/Agent', () => {
       expect(agentAfterDelete).toBeNull();
     });
 
+    test('should remove ACL entries when deleting an agent', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Agent With Permissions',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+      });
+
+      // Grant permissions (simulating sharing)
+      await permissionService.grantPermission({
+        principalType: PrincipalType.USER,
+        principalId: authorId,
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+        accessRoleId: AccessRoleIds.AGENT_OWNER,
+        grantedBy: authorId,
+      });
+
+      // Verify ACL entry exists
+      const aclEntriesBefore = await AclEntry.find({
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+      });
+      expect(aclEntriesBefore).toHaveLength(1);
+
+      // Delete the agent
+      await deleteAgent({ id: agentId });
+
+      // Verify agent is deleted
+      const agentAfterDelete = await getAgent({ id: agentId });
+      expect(agentAfterDelete).toBeNull();
+
+      // Verify ACL entries are removed
+      const aclEntriesAfter = await AclEntry.find({
+        resourceType: ResourceType.AGENT,
+        resourceId: agent._id,
+      });
+      expect(aclEntriesAfter).toHaveLength(0);
+    });
+
     test('should list agents by author', async () => {
       const authorId = new mongoose.Types.ObjectId();
       const otherAuthorId = new mongoose.Types.ObjectId();
@@ -1258,6 +1321,328 @@ describe('models/Agent', () => {
       expect(secondUpdate.versions).toHaveLength(3);
     });
 
+    test('should detect changes in support_contact fields', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent with initial support_contact
+      await createAgent({
+        id: agentId,
+        name: 'Agent with Support Contact',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+        support_contact: {
+          name: 'Initial Support',
+          email: 'initial@support.com',
+        },
+      });
+
+      // Update support_contact name only
+      const firstUpdate = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'Updated Support',
+            email: 'initial@support.com',
+          },
+        },
+      );
+
+      expect(firstUpdate.versions).toHaveLength(2);
+      expect(firstUpdate.support_contact.name).toBe('Updated Support');
+      expect(firstUpdate.support_contact.email).toBe('initial@support.com');
+
+      // Update support_contact email only
+      const secondUpdate = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'Updated Support',
+            email: 'updated@support.com',
+          },
+        },
+      );
+
+      expect(secondUpdate.versions).toHaveLength(3);
+      expect(secondUpdate.support_contact.email).toBe('updated@support.com');
+
+      // Try to update with same support_contact - should be detected as duplicate
+      await expect(
+        updateAgent(
+          { id: agentId },
+          {
+            support_contact: {
+              name: 'Updated Support',
+              email: 'updated@support.com',
+            },
+          },
+        ),
+      ).rejects.toThrow('Duplicate version');
+    });
+
+    test('should handle support_contact from empty to populated', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent without support_contact
+      const agent = await createAgent({
+        id: agentId,
+        name: 'Agent without Support',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+      });
+
+      // Verify support_contact is undefined since it wasn't provided
+      expect(agent.support_contact).toBeUndefined();
+
+      // Update to add support_contact
+      const updated = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'New Support Team',
+            email: 'support@example.com',
+          },
+        },
+      );
+
+      expect(updated.versions).toHaveLength(2);
+      expect(updated.support_contact.name).toBe('New Support Team');
+      expect(updated.support_contact.email).toBe('support@example.com');
+    });
+
+    test('should handle support_contact edge cases in isDuplicateVersion', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent with support_contact
+      await createAgent({
+        id: agentId,
+        name: 'Edge Case Agent',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+        support_contact: {
+          name: 'Support',
+          email: 'support@test.com',
+        },
+      });
+
+      // Update to empty support_contact
+      const emptyUpdate = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {},
+        },
+      );
+
+      expect(emptyUpdate.versions).toHaveLength(2);
+      expect(emptyUpdate.support_contact).toEqual({});
+
+      // Update back to populated support_contact
+      const repopulated = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'Support',
+            email: 'support@test.com',
+          },
+        },
+      );
+
+      expect(repopulated.versions).toHaveLength(3);
+
+      // Verify all versions have correct support_contact
+      const finalAgent = await getAgent({ id: agentId });
+      expect(finalAgent.versions[0].support_contact).toEqual({
+        name: 'Support',
+        email: 'support@test.com',
+      });
+      expect(finalAgent.versions[1].support_contact).toEqual({});
+      expect(finalAgent.versions[2].support_contact).toEqual({
+        name: 'Support',
+        email: 'support@test.com',
+      });
+    });
+
+    test('should preserve support_contact in version history', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent
+      await createAgent({
+        id: agentId,
+        name: 'Version History Test',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+        support_contact: {
+          name: 'Initial Contact',
+          email: 'initial@test.com',
+        },
+      });
+
+      // Multiple updates with different support_contact values
+      await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'Second Contact',
+            email: 'second@test.com',
+          },
+        },
+      );
+
+      await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'Third Contact',
+            email: 'third@test.com',
+          },
+        },
+      );
+
+      const finalAgent = await getAgent({ id: agentId });
+
+      // Verify version history
+      expect(finalAgent.versions).toHaveLength(3);
+      expect(finalAgent.versions[0].support_contact).toEqual({
+        name: 'Initial Contact',
+        email: 'initial@test.com',
+      });
+      expect(finalAgent.versions[1].support_contact).toEqual({
+        name: 'Second Contact',
+        email: 'second@test.com',
+      });
+      expect(finalAgent.versions[2].support_contact).toEqual({
+        name: 'Third Contact',
+        email: 'third@test.com',
+      });
+
+      // Current state should match last version
+      expect(finalAgent.support_contact).toEqual({
+        name: 'Third Contact',
+        email: 'third@test.com',
+      });
+    });
+
+    test('should handle partial support_contact updates', async () => {
+      const agentId = `agent_${uuidv4()}`;
+      const authorId = new mongoose.Types.ObjectId();
+
+      // Create agent with full support_contact
+      await createAgent({
+        id: agentId,
+        name: 'Partial Update Test',
+        provider: 'test',
+        model: 'test-model',
+        author: authorId,
+        support_contact: {
+          name: 'Original Name',
+          email: 'original@email.com',
+        },
+      });
+
+      // MongoDB's findOneAndUpdate will replace the entire support_contact object
+      // So we need to verify that partial updates still work correctly
+      const updated = await updateAgent(
+        { id: agentId },
+        {
+          support_contact: {
+            name: 'New Name',
+            email: '', // Empty email
+          },
+        },
+      );
+
+      expect(updated.versions).toHaveLength(2);
+      expect(updated.support_contact.name).toBe('New Name');
+      expect(updated.support_contact.email).toBe('');
+
+      // Verify isDuplicateVersion works with partial changes
+      await expect(
+        updateAgent(
+          { id: agentId },
+          {
+            support_contact: {
+              name: 'New Name',
+              email: '',
+            },
+          },
+        ),
+      ).rejects.toThrow('Duplicate version');
+    });
+
+    // Edge Cases
+    describe.each([
+      {
+        operation: 'add',
+        name: 'empty file_id',
+        needsAgent: true,
+        params: { tool_resource: 'file_search', file_id: '' },
+        shouldResolve: true,
+      },
+      {
+        operation: 'add',
+        name: 'non-existent agent',
+        needsAgent: false,
+        params: { tool_resource: 'file_search', file_id: 'file123' },
+        shouldResolve: false,
+        error: 'Agent not found for adding resource file',
+      },
+    ])('addAgentResourceFile with $name', ({ needsAgent, params, shouldResolve, error }) => {
+      test(`should ${shouldResolve ? 'resolve' : 'reject'}`, async () => {
+        const agent = needsAgent ? await createBasicAgent() : null;
+        const agent_id = needsAgent ? agent.id : `agent_${uuidv4()}`;
+
+        if (shouldResolve) {
+          await expect(addAgentResourceFile({ agent_id, ...params })).resolves.toBeDefined();
+        } else {
+          await expect(addAgentResourceFile({ agent_id, ...params })).rejects.toThrow(error);
+        }
+      });
+    });
+
+    describe.each([
+      {
+        name: 'empty files array',
+        files: [],
+        needsAgent: true,
+        shouldResolve: true,
+      },
+      {
+        name: 'non-existent tool_resource',
+        files: [{ tool_resource: 'non_existent_tool', file_id: 'file123' }],
+        needsAgent: true,
+        shouldResolve: true,
+      },
+      {
+        name: 'non-existent agent',
+        files: [{ tool_resource: 'file_search', file_id: 'file123' }],
+        needsAgent: false,
+        shouldResolve: false,
+        error: 'Agent not found for removing resource files',
+      },
+    ])('removeAgentResourceFiles with $name', ({ files, needsAgent, shouldResolve, error }) => {
+      test(`should ${shouldResolve ? 'resolve' : 'reject'}`, async () => {
+        const agent = needsAgent ? await createBasicAgent() : null;
+        const agent_id = needsAgent ? agent.id : `agent_${uuidv4()}`;
+
+        if (shouldResolve) {
+          const result = await removeAgentResourceFiles({ agent_id, files });
+          expect(result).toBeDefined();
+          if (agent) {
+            expect(result.id).toBe(agent.id);
+          }
+        } else {
+          await expect(removeAgentResourceFiles({ agent_id, files })).rejects.toThrow(error);
+        }
+      });
+    });
+
     describe('Edge Cases', () => {
       test('should handle extremely large version history', async () => {
         const agentId = `agent_${uuidv4()}`;
@@ -2578,6 +2963,299 @@ describe('models/Agent', () => {
   });
 });
 
+describe('Support Contact Field', () => {
+  let mongoServer;
+
+  beforeAll(async () => {
+    mongoServer = await MongoMemoryServer.create();
+    const mongoUri = mongoServer.getUri();
+    Agent = mongoose.models.Agent || mongoose.model('Agent', agentSchema);
+    await mongoose.connect(mongoUri);
+  }, 20000);
+
+  afterAll(async () => {
+    await mongoose.disconnect();
+    await mongoServer.stop();
+  });
+
+  beforeEach(async () => {
+    await Agent.deleteMany({});
+  });
+
+  it('should not create subdocument with ObjectId for support_contact', async () => {
+    const userId = new mongoose.Types.ObjectId();
+    const agentData = {
+      id: 'agent_test_support',
+      name: 'Test Agent',
+      provider: 'openai',
+      model: 'gpt-4',
+      author: userId,
+      support_contact: {
+        name: 'Support Team',
+        email: 'support@example.com',
+      },
+    };
+
+    // Create agent
+    const agent = await createAgent(agentData);
+
+    // Verify support_contact is stored correctly
+    expect(agent.support_contact).toBeDefined();
+    expect(agent.support_contact.name).toBe('Support Team');
+    expect(agent.support_contact.email).toBe('support@example.com');
+
+    // Verify no _id field is created in support_contact
+    expect(agent.support_contact._id).toBeUndefined();
+
+    // Fetch from database to double-check
+    const dbAgent = await Agent.findOne({ id: agentData.id });
+    expect(dbAgent.support_contact).toBeDefined();
+    expect(dbAgent.support_contact.name).toBe('Support Team');
+    expect(dbAgent.support_contact.email).toBe('support@example.com');
+    expect(dbAgent.support_contact._id).toBeUndefined();
+  });
+
+  it('should handle empty support_contact correctly', async () => {
+    const userId = new mongoose.Types.ObjectId();
+    const agentData = {
+      id: 'agent_test_empty_support',
+      name: 'Test Agent',
+      provider: 'openai',
+      model: 'gpt-4',
+      author: userId,
+      support_contact: {},
+    };
+
+    const agent = await createAgent(agentData);
+
+    // Verify empty support_contact is stored as empty object
+    expect(agent.support_contact).toEqual({});
+    expect(agent.support_contact._id).toBeUndefined();
+  });
+
+  it('should handle missing support_contact correctly', async () => {
+    const userId = new mongoose.Types.ObjectId();
+    const agentData = {
+      id: 'agent_test_no_support',
+      name: 'Test Agent',
+      provider: 'openai',
+      model: 'gpt-4',
+      author: userId,
+    };
+
+    const agent = await createAgent(agentData);
+
+    // Verify support_contact is undefined when not provided
+    expect(agent.support_contact).toBeUndefined();
+  });
+
+  describe('getListAgentsByAccess - Security Tests', () => {
+    let userA, userB;
+    let agentA1, agentA2, agentA3;
+
+    beforeEach(async () => {
+      Agent = mongoose.models.Agent || mongoose.model('Agent', agentSchema);
+      await Agent.deleteMany({});
+      await AclEntry.deleteMany({});
+
+      // Create two users
+      userA = new mongoose.Types.ObjectId();
+      userB = new mongoose.Types.ObjectId();
+
+      // Create agents for user A
+      agentA1 = await createAgent({
+        id: `agent_${uuidv4().slice(0, 12)}`,
+        name: 'Agent A1',
+        description: 'User A agent 1',
+        provider: 'openai',
+        model: 'gpt-4',
+        author: userA,
+      });
+
+      agentA2 = await createAgent({
+        id: `agent_${uuidv4().slice(0, 12)}`,
+        name: 'Agent A2',
+        description: 'User A agent 2',
+        provider: 'openai',
+        model: 'gpt-4',
+        author: userA,
+      });
+
+      agentA3 = await createAgent({
+        id: `agent_${uuidv4().slice(0, 12)}`,
+        name: 'Agent A3',
+        description: 'User A agent 3',
+        provider: 'openai',
+        model: 'gpt-4',
+        author: userA,
+      });
+    });
+
+    test('should return empty list when user has no accessible agents (empty accessibleIds)', async () => {
+      // User B has no agents and no shared agents
+      const result = await getListAgentsByAccess({
+        accessibleIds: [],
+        otherParams: {},
+      });
+
+      expect(result.data).toHaveLength(0);
+      expect(result.has_more).toBe(false);
+      expect(result.first_id).toBeNull();
+      expect(result.last_id).toBeNull();
+    });
+
+    test('should not return other users agents when accessibleIds is empty', async () => {
+      // User B trying to list agents with empty accessibleIds should not see User A's agents
+      const result = await getListAgentsByAccess({
+        accessibleIds: [],
+        otherParams: { author: userB },
+      });
+
+      expect(result.data).toHaveLength(0);
+      expect(result.has_more).toBe(false);
+    });
+
+    test('should only return agents in accessibleIds list', async () => {
+      // Give User B access to only one of User A's agents
+      const accessibleIds = [agentA1._id];
+
+      const result = await getListAgentsByAccess({
+        accessibleIds,
+        otherParams: {},
+      });
+
+      expect(result.data).toHaveLength(1);
+      expect(result.data[0].id).toBe(agentA1.id);
+      expect(result.data[0].name).toBe('Agent A1');
+    });
+
+    test('should return multiple accessible agents when provided', async () => {
+      // Give User B access to two of User A's agents
+      const accessibleIds = [agentA1._id, agentA3._id];
+
+      const result = await getListAgentsByAccess({
+        accessibleIds,
+        otherParams: {},
+      });
+
+      expect(result.data).toHaveLength(2);
+      const returnedIds = result.data.map((agent) => agent.id);
+      expect(returnedIds).toContain(agentA1.id);
+      expect(returnedIds).toContain(agentA3.id);
+      expect(returnedIds).not.toContain(agentA2.id);
+    });
+
+    test('should respect other query parameters while enforcing accessibleIds', async () => {
+      // Give access to all agents but filter by name
+      const accessibleIds = [agentA1._id, agentA2._id, agentA3._id];
+
+      const result = await getListAgentsByAccess({
+        accessibleIds,
+        otherParams: { name: 'Agent A2' },
+      });
+
+      expect(result.data).toHaveLength(1);
+      expect(result.data[0].id).toBe(agentA2.id);
+    });
+
+    test('should handle pagination correctly with accessibleIds filter', async () => {
+      // Create more agents
+      const moreAgents = [];
+      for (let i = 4; i <= 10; i++) {
+        const agent = await createAgent({
+          id: `agent_${uuidv4().slice(0, 12)}`,
+          name: `Agent A${i}`,
+          description: `User A agent ${i}`,
+          provider: 'openai',
+          model: 'gpt-4',
+          author: userA,
+        });
+        moreAgents.push(agent);
+      }
+
+      // Give access to all agents
+      const allAgentIds = [agentA1, agentA2, agentA3, ...moreAgents].map((a) => a._id);
+
+      // First page
+      const page1 = await getListAgentsByAccess({
+        accessibleIds: allAgentIds,
+        otherParams: {},
+        limit: 5,
+      });
+
+      expect(page1.data).toHaveLength(5);
+      expect(page1.has_more).toBe(true);
+      expect(page1.after).toBeTruthy();
+
+      // Second page
+      const page2 = await getListAgentsByAccess({
+        accessibleIds: allAgentIds,
+        otherParams: {},
+        limit: 5,
+        after: page1.after,
+      });
+
+      expect(page2.data).toHaveLength(5);
+      expect(page2.has_more).toBe(false);
+
+      // Verify no overlap between pages
+      const page1Ids = page1.data.map((a) => a.id);
+      const page2Ids = page2.data.map((a) => a.id);
+      const intersection = page1Ids.filter((id) => page2Ids.includes(id));
+      expect(intersection).toHaveLength(0);
+    });
+
+    test('should return empty list when accessibleIds contains non-existent IDs', async () => {
+      // Try with non-existent agent IDs
+      const fakeIds = [new mongoose.Types.ObjectId(), new mongoose.Types.ObjectId()];
+
+      const result = await getListAgentsByAccess({
+        accessibleIds: fakeIds,
+        otherParams: {},
+      });
+
+      expect(result.data).toHaveLength(0);
+      expect(result.has_more).toBe(false);
+    });
+
+    test('should handle undefined accessibleIds as empty array', async () => {
+      // When accessibleIds is undefined, it should be treated as empty array
+      const result = await getListAgentsByAccess({
+        accessibleIds: undefined,
+        otherParams: {},
+      });
+
+      expect(result.data).toHaveLength(0);
+      expect(result.has_more).toBe(false);
+    });
+
+    test('should combine accessibleIds with author filter correctly', async () => {
+      // Create an agent for User B
+      const agentB1 = await createAgent({
+        id: `agent_${uuidv4().slice(0, 12)}`,
+        name: 'Agent B1',
+        description: 'User B agent 1',
+        provider: 'openai',
+        model: 'gpt-4',
+        author: userB,
+      });
+
+      // Give User B access to one of User A's agents
+      const accessibleIds = [agentA1._id, agentB1._id];
+
+      // Filter by author should further restrict the results
+      const result = await getListAgentsByAccess({
+        accessibleIds,
+        otherParams: { author: userB },
+      });
+
+      expect(result.data).toHaveLength(1);
+      expect(result.data[0].id).toBe(agentB1.id);
+      expect(result.data[0].author).toBe(userB.toString());
+    });
+  });
+});
+
 function createBasicAgent(overrides = {}) {
   const defaults = {
     id: `agent_${uuidv4()}`,

api/models/Conversation.js

@@ -1,4 +1,6 @@
 const { logger } = require('@librechat/data-schemas');
+const { createTempChatExpirationDate } = require('@librechat/api');
+const { getCustomConfig } = require('~/server/services/Config/getCustomConfig');
 const { getMessages, deleteMessages } = require('./Message');
 const { Conversation } = require('~/db/models');
 
@@ -98,10 +100,15 @@ 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 customConfig = await getCustomConfig();
+          update.expiredAt = createTempChatExpirationDate(customConfig);
+        } 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 +1,572 @@
+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/getCustomConfig');
+jest.mock('./Message');
+const { getCustomConfig } = require('~/server/services/Config/getCustomConfig');
+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: {},
+    };
+
+    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 custom config with 24 hour retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with 48 hour retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with less than minimum retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with more than maximum retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 getCustomConfig errors gracefully', async () => {
+      // Mock getCustomConfig to throw an error
+      getCustomConfig.mockRejectedValue(new Error('Config service unavailable'));
+
+      mockReq.body = { isTemporary: true };
+
+      const result = await saveConvo(mockReq, mockConversationData);
+
+      // Should still save the conversation but with expiredAt as null
+      expect(result.conversationId).toBe(mockConversationData.conversationId);
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should use default retention when config is not provided', async () => {
+      // Mock getCustomConfig to return empty config
+      getCustomConfig.mockResolvedValue({});
+
+      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
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 +1,562 @@
+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', 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,
+      });
+
+      // 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);
+    });
+  });
+
+  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 +1,7 @@
 const { z } = require('zod');
 const { logger } = require('@librechat/data-schemas');
+const { createTempChatExpirationDate } = require('@librechat/api');
+const { getCustomConfig } = require('~/server/services/Config/getCustomConfig');
 const { Message } = require('~/db/models');
 
 const idSchema = z.string().uuid();
@@ -54,9 +56,14 @@ async function saveMessage(req, params, metadata) {
     };
 
     if (req?.body?.isTemporary) {
-      const expiredAt = new Date();
-      expiredAt.setDate(expiredAt.getDate() + 30);
-      update.expiredAt = expiredAt;
+      try {
+        const customConfig = await getCustomConfig();
+        update.expiredAt = createTempChatExpirationDate(customConfig);
+      } 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 +1,21 @@
 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/getCustomConfig');
+const { getCustomConfig } = require('~/server/services/Config/getCustomConfig');
+
 /**
  * @type {import('mongoose').Model<import('@librechat/data-schemas').IMessage>}
  */
@@ -117,21 +121,21 @@ 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 +318,265 @@ 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 custom config with 24 hour retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with 48 hour retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with less than minimum retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 custom config with more than maximum retention
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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 getCustomConfig errors gracefully', async () => {
+      // Mock getCustomConfig to throw an error
+      getCustomConfig.mockRejectedValue(new Error('Config service unavailable'));
+
+      mockReq.body = { isTemporary: true };
+
+      const result = await saveMessage(mockReq, mockMessageData);
+
+      // Should still save the message but with expiredAt as null
+      expect(result.messageId).toBe('msg123');
+      expect(result.expiredAt).toBeNull();
+    });
+
+    it('should use default retention when config is not provided', async () => {
+      // Mock getCustomConfig to return empty config
+      getCustomConfig.mockResolvedValue({});
+
+      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
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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
+      getCustomConfig.mockResolvedValue({
+        interface: {
+          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) {
@@ -235,6 +237,13 @@ 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' };
 };
 
@@ -430,6 +439,16 @@ 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');

api/models/index.js

@@ -22,6 +22,7 @@ const {
 } = require('./Message');
 const { getConvoTitle, getConvo, saveConvo, deleteConvos } = require('./Conversation');
 const { getPreset, getPresets, savePreset, deletePresets } = require('./Preset');
+const { File } = require('~/db/models');
 
 module.exports = {
   ...methods,
@@ -51,4 +52,6 @@ module.exports = {
   getPresets,
   savePreset,
   deletePresets,
+
+  Files: File,
 };

api/models/tx.js

@@ -135,10 +135,11 @@ 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 },

api/models/tx.spec.js

@@ -636,6 +636,15 @@ 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 +671,15 @@ 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/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@librechat/backend",
-  "version": "v0.7.8",
+  "version": "v0.8.0-rc1",
   "description": "",
   "scripts": {
     "start": "echo 'please run this from the root directory'",
@@ -44,21 +44,22 @@
     "@googleapis/youtube": "^20.0.0",
     "@keyv/redis": "^4.3.3",
     "@langchain/community": "^0.3.47",
-    "@langchain/core": "^0.3.60",
+    "@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": "^2.4.41",
+    "@librechat/agents": "^2.4.69",
     "@librechat/api": "*",
     "@librechat/data-schemas": "*",
-    "@node-saml/passport-saml": "^5.0.0",
+    "@modelcontextprotocol/sdk": "^1.17.1",
+    "@node-saml/passport-saml": "^5.1.0",
     "@microsoft/microsoft-graph-client": "^3.0.7",
     "@waylaidwanderer/fetch-event-source": "^3.0.1",
     "axios": "^1.8.2",
     "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",
@@ -68,10 +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",
@@ -89,12 +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": "^5.10.1",
     "openai-chat-tokens": "^0.2.8",
     "openid-client": "^6.5.0",
     "passport": "^0.6.0",
@@ -119,7 +121,7 @@
   },
   "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

@@ -169,9 +169,6 @@ function disposeClient(client) {
       client.isGenerativeModel = null;
     }
     // Properties specific to OpenAIClient
-    if (client.ChatGPTClient) {
-      client.ChatGPTClient = null;
-    }
     if (client.completionsUrl) {
       client.completionsUrl = null;
     }

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 {
@@ -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 {
@@ -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

@@ -24,17 +24,23 @@ const handleValidationError = (err, res) => {
   }
 };
 
-// eslint-disable-next-line no-unused-vars
-module.exports = (err, req, res, next) => {
+module.exports = (err, _req, res, _next) => {
   try {
     if (err.name === 'ValidationError') {
-      return (err = handleValidationError(err, res));
+      return handleValidationError(err, res);
     }
     if (err.code && err.code == 11000) {
-      return (err = handleDuplicateKeyError(err, res));
+      return handleDuplicateKeyError(err, res);
     }
-  } catch (err) {
+    // Special handling for errors like SyntaxError
+    if (err.statusCode && err.body) {
+      return res.status(err.statusCode).send(err.body);
+    }
+
     logger.error('ErrorController => error', err);
-    res.status(500).send('An unknown error occurred.');
+    return res.status(500).send('An unknown error occurred.');
+  } catch (err) {
+    logger.error('ErrorController => processing error', err);
+    return res.status(500).send('Processing error in ErrorController.');
   }
 };

api/server/controllers/ErrorController.spec.js

@@ -0,0 +1,241 @@
+const errorController = require('./ErrorController');
+const { logger } = require('~/config');
+
+// Mock the logger
+jest.mock('~/config', () => ({
+  logger: {
+    error: jest.fn(),
+  },
+}));
+
+describe('ErrorController', () => {
+  let mockReq, mockRes, mockNext;
+
+  beforeEach(() => {
+    mockReq = {};
+    mockRes = {
+      status: jest.fn().mockReturnThis(),
+      send: jest.fn(),
+    };
+    mockNext = jest.fn();
+    logger.error.mockClear();
+  });
+
+  describe('ValidationError handling', () => {
+    it('should handle ValidationError with single error', () => {
+      const validationError = {
+        name: 'ValidationError',
+        errors: {
+          email: { message: 'Email is required', path: 'email' },
+        },
+      };
+
+      errorController(validationError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(400);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: '["Email is required"]',
+        fields: '["email"]',
+      });
+      expect(logger.error).toHaveBeenCalledWith('Validation error:', validationError.errors);
+    });
+
+    it('should handle ValidationError with multiple errors', () => {
+      const validationError = {
+        name: 'ValidationError',
+        errors: {
+          email: { message: 'Email is required', path: 'email' },
+          password: { message: 'Password is required', path: 'password' },
+        },
+      };
+
+      errorController(validationError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(400);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: '"Email is required Password is required"',
+        fields: '["email","password"]',
+      });
+      expect(logger.error).toHaveBeenCalledWith('Validation error:', validationError.errors);
+    });
+
+    it('should handle ValidationError with empty errors object', () => {
+      const validationError = {
+        name: 'ValidationError',
+        errors: {},
+      };
+
+      errorController(validationError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(400);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: '[]',
+        fields: '[]',
+      });
+    });
+  });
+
+  describe('Duplicate key error handling', () => {
+    it('should handle duplicate key error (code 11000)', () => {
+      const duplicateKeyError = {
+        code: 11000,
+        keyValue: { email: 'test@example.com' },
+      };
+
+      errorController(duplicateKeyError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(409);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: 'An document with that ["email"] already exists.',
+        fields: '["email"]',
+      });
+      expect(logger.error).toHaveBeenCalledWith('Duplicate key error:', duplicateKeyError.keyValue);
+    });
+
+    it('should handle duplicate key error with multiple fields', () => {
+      const duplicateKeyError = {
+        code: 11000,
+        keyValue: { email: 'test@example.com', username: 'testuser' },
+      };
+
+      errorController(duplicateKeyError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(409);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: 'An document with that ["email","username"] already exists.',
+        fields: '["email","username"]',
+      });
+      expect(logger.error).toHaveBeenCalledWith('Duplicate key error:', duplicateKeyError.keyValue);
+    });
+
+    it('should handle error with code 11000 as string', () => {
+      const duplicateKeyError = {
+        code: '11000',
+        keyValue: { email: 'test@example.com' },
+      };
+
+      errorController(duplicateKeyError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(409);
+      expect(mockRes.send).toHaveBeenCalledWith({
+        messages: 'An document with that ["email"] already exists.',
+        fields: '["email"]',
+      });
+    });
+  });
+
+  describe('SyntaxError handling', () => {
+    it('should handle errors with statusCode and body', () => {
+      const syntaxError = {
+        statusCode: 400,
+        body: 'Invalid JSON syntax',
+      };
+
+      errorController(syntaxError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(400);
+      expect(mockRes.send).toHaveBeenCalledWith('Invalid JSON syntax');
+    });
+
+    it('should handle errors with different statusCode and body', () => {
+      const customError = {
+        statusCode: 422,
+        body: { error: 'Unprocessable entity' },
+      };
+
+      errorController(customError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(422);
+      expect(mockRes.send).toHaveBeenCalledWith({ error: 'Unprocessable entity' });
+    });
+
+    it('should handle error with statusCode but no body', () => {
+      const partialError = {
+        statusCode: 400,
+      };
+
+      errorController(partialError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.send).toHaveBeenCalledWith('An unknown error occurred.');
+    });
+
+    it('should handle error with body but no statusCode', () => {
+      const partialError = {
+        body: 'Some error message',
+      };
+
+      errorController(partialError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.send).toHaveBeenCalledWith('An unknown error occurred.');
+    });
+  });
+
+  describe('Unknown error handling', () => {
+    it('should handle unknown errors', () => {
+      const unknownError = new Error('Some unknown error');
+
+      errorController(unknownError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.send).toHaveBeenCalledWith('An unknown error occurred.');
+      expect(logger.error).toHaveBeenCalledWith('ErrorController => error', unknownError);
+    });
+
+    it('should handle errors with code other than 11000', () => {
+      const mongoError = {
+        code: 11100,
+        message: 'Some MongoDB error',
+      };
+
+      errorController(mongoError, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.send).toHaveBeenCalledWith('An unknown error occurred.');
+      expect(logger.error).toHaveBeenCalledWith('ErrorController => error', mongoError);
+    });
+
+    it('should handle null/undefined errors', () => {
+      errorController(null, mockReq, mockRes, mockNext);
+
+      expect(mockRes.status).toHaveBeenCalledWith(500);
+      expect(mockRes.send).toHaveBeenCalledWith('Processing error in ErrorController.');
+      expect(logger.error).toHaveBeenCalledWith(
+        'ErrorController => processing error',
+        expect.any(Error),
+      );
+    });
+  });
+
+  describe('Catch block handling', () => {
+    beforeEach(() => {
+      // Restore logger mock to normal behavior for these tests
+      logger.error.mockRestore();
+      logger.error = jest.fn();
+    });
+
+    it('should handle errors when logger.error throws', () => {
+      // Create fresh mocks for this test
+      const freshMockRes = {
+        status: jest.fn().mockReturnThis(),
+        send: jest.fn(),
+      };
+
+      // Mock logger to throw on the first call, succeed on the second
+      logger.error
+        .mockImplementationOnce(() => {
+          throw new Error('Logger error');
+        })
+        .mockImplementation(() => {});
+
+      const testError = new Error('Test error');
+
+      errorController(testError, mockReq, freshMockRes, mockNext);
+
+      expect(freshMockRes.status).toHaveBeenCalledWith(500);
+      expect(freshMockRes.send).toHaveBeenCalledWith('Processing error in ErrorController.');
+      expect(logger.error).toHaveBeenCalledTimes(2);
+    });
+  });
+});

api/server/controllers/PermissionsController.js

@@ -4,12 +4,13 @@
 
 const mongoose = require('mongoose');
 const { logger } = require('@librechat/data-schemas');
+const { ResourceType, PrincipalType } = require('librechat-data-provider');
 const {
-  getAvailableRoles,
-  ensurePrincipalExists,
-  getEffectivePermissions,
-  ensureGroupPrincipalExists,
   bulkUpdateResourcePermissions,
+  ensureGroupPrincipalExists,
+  getEffectivePermissions,
+  ensurePrincipalExists,
+  getAvailableRoles,
 } = require('~/server/services/PermissionService');
 const { AclEntry } = require('~/db/models');
 const {
@@ -18,8 +19,8 @@ const {
   calculateRelevanceScore,
 } = require('~/models');
 const {
-  searchEntraIdPrincipals,
   entraIdPrincipalFeatureEnabled,
+  searchEntraIdPrincipals,
 } = require('~/server/services/GraphApiService');
 
 /**
@@ -27,6 +28,18 @@ const {
  * 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
@@ -41,6 +54,8 @@ const {
 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;
@@ -57,7 +72,7 @@ const updateResourcePermissions = async (req, res) => {
     // Add public permission if enabled
     if (isPublic && publicAccessRoleId) {
       updatedPrincipals.push({
-        type: 'public',
+        type: PrincipalType.PUBLIC,
         id: null,
         accessRoleId: publicAccessRoleId,
       });
@@ -82,11 +97,13 @@ const updateResourcePermissions = async (req, res) => {
       try {
         let principalId;
 
-        if (principal.type === 'public') {
+        if (principal.type === PrincipalType.PUBLIC) {
           principalId = null; // Public principals don't need database records
-        } else if (principal.type === 'user') {
+        } 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 === 'group') {
+        } else if (principal.type === PrincipalType.GROUP) {
           // Pass authContext to enable member fetching for Entra ID groups when available
           principalId = await ensureGroupPrincipalExists(principal, authContext);
         } else {
@@ -122,7 +139,7 @@ const updateResourcePermissions = async (req, res) => {
     // If public is disabled, add public to revoked list
     if (!isPublic) {
       revokedPrincipals.push({
-        type: 'public',
+        type: PrincipalType.PUBLIC,
         id: null,
       });
     }
@@ -163,6 +180,7 @@ const updateResourcePermissions = async (req, res) => {
 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([
@@ -219,14 +237,14 @@ const getResourcePermissions = async (req, res) => {
 
     // Process aggregation results
     for (const result of results) {
-      if (result.principalType === 'public') {
+      if (result.principalType === PrincipalType.PUBLIC) {
         publicPermission = {
           public: true,
           publicAccessRoleId: result.accessRoleId,
         };
-      } else if (result.principalType === 'user' && result.userInfo) {
+      } else if (result.principalType === PrincipalType.USER && result.userInfo) {
         principals.push({
-          type: 'user',
+          type: PrincipalType.USER,
           id: result.userInfo._id.toString(),
           name: result.userInfo.name || result.userInfo.username,
           email: result.userInfo.email,
@@ -235,9 +253,9 @@ const getResourcePermissions = async (req, res) => {
           idOnTheSource: result.userInfo.idOnTheSource || result.userInfo._id.toString(),
           accessRoleId: result.accessRoleId,
         });
-      } else if (result.principalType === 'group' && result.groupInfo) {
+      } else if (result.principalType === PrincipalType.GROUP && result.groupInfo) {
         principals.push({
-          type: 'group',
+          type: PrincipalType.GROUP,
           id: result.groupInfo._id.toString(),
           name: result.groupInfo.name,
           email: result.groupInfo.email,
@@ -247,6 +265,16 @@ const getResourcePermissions = async (req, res) => {
           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,
+        });
       }
     }
 
@@ -278,6 +306,7 @@ const getResourcePermissions = async (req, res) => {
 const getResourceRoles = async (req, res) => {
   try {
     const { resourceType } = req.params;
+    validateResourceType(resourceType);
 
     const roles = await getAvailableRoles({ resourceType });
 
@@ -305,10 +334,13 @@ const getResourceRoles = async (req, res) => {
 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,
     });
@@ -347,7 +379,9 @@ const searchPrincipals = async (req, res) => {
     }
 
     const searchLimit = Math.min(Math.max(1, parseInt(limit) || 10), 50);
-    const typeFilter = ['user', 'group'].includes(type) ? type : null;
+    const typeFilter = [PrincipalType.USER, PrincipalType.GROUP, PrincipalType.ROLE].includes(type)
+      ? type
+      : null;
 
     const localResults = await searchLocalPrincipals(query.trim(), searchLimit, typeFilter);
     let allPrincipals = [...localResults];

api/server/controllers/PluginController.js

@@ -1,11 +1,10 @@
 const { logger } = require('@librechat/data-schemas');
-const { CacheKeys, AuthType } = require('librechat-data-provider');
+const { CacheKeys, AuthType, Constants } = require('librechat-data-provider');
 const { getCustomConfig, getCachedTools } = require('~/server/services/Config');
 const { getToolkitKey } = require('~/server/services/ToolService');
 const { getMCPManager, getFlowStateManager } = require('~/config');
 const { availableTools } = require('~/app/clients/tools');
 const { getLogStores } = require('~/cache');
-const { Constants } = require('librechat-data-provider');
 
 /**
  * Filters out duplicate plugins from the list of plugins.
@@ -139,15 +138,21 @@ function createGetServerTools() {
  */
 const getAvailableTools = async (req, res) => {
   try {
+    const userId = req.user?.id;
+    const customConfig = await getCustomConfig();
     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 userPlugins = convertMCPToolsToPlugins(cachedUserTools, customConfig);
+
+    if (cachedToolsArray && userPlugins) {
+      const dedupedTools = filterUniquePlugins([...userPlugins, ...cachedToolsArray]);
+      res.status(200).json(dedupedTools);
       return;
     }
 
+    // If not in cache, build from manifest
     let pluginManifest = availableTools;
-    const customConfig = await getCustomConfig();
     if (customConfig?.mcpServers != null) {
       const mcpManager = getMCPManager();
       const flowsCache = getLogStores(CacheKeys.FLOWS);
@@ -173,7 +178,7 @@ const getAvailableTools = async (req, res) => {
       }
     });
 
-    const toolDefinitions = await getCachedTools({ includeGlobal: true });
+    const toolDefinitions = (await getCachedTools({ includeGlobal: true })) || {};
 
     const toolsOutput = [];
     for (const plugin of authenticatedPlugins) {
@@ -218,16 +223,70 @@ const getAvailableTools = async (req, res) => {
 
       toolsOutput.push(toolToAdd);
     }
-
     const finalTools = filterUniquePlugins(toolsOutput);
     await cache.set(CacheKeys.TOOLS, finalTools);
-    res.status(200).json(finalTools);
+
+    const dedupedTools = filterUniquePlugins([...userPlugins, ...finalTools]);
+
+    res.status(200).json(dedupedTools);
   } catch (error) {
     logger.error('[getAvailableTools]', error);
     res.status(500).json({ message: error.message });
   }
 };
 
+/**
+ * Converts MCP function format tools to plugin format
+ * @param {Object} functionTools - Object with function format tools
+ * @param {Object} customConfig - Custom configuration for MCP servers
+ * @returns {Array} Array of plugin objects
+ */
+function convertMCPToolsToPlugins(functionTools, customConfig) {
+  const plugins = [];
+
+  for (const [toolKey, toolData] of Object.entries(functionTools)) {
+    if (!toolData.function || !toolKey.includes(Constants.mcp_delimiter)) {
+      continue;
+    }
+
+    const functionData = toolData.function;
+    const parts = toolKey.split(Constants.mcp_delimiter);
+    const serverName = parts[parts.length - 1];
+
+    const serverConfig = customConfig?.mcpServers?.[serverName];
+
+    const plugin = {
+      name: parts[0], // Use the tool name without server suffix
+      pluginKey: toolKey,
+      description: functionData.description || '',
+      authenticated: true,
+      icon: serverConfig?.iconPath,
+    };
+
+    // Build authConfig for MCP tools
+    if (!serverConfig?.customUserVars) {
+      plugin.authConfig = [];
+      plugins.push(plugin);
+      continue;
+    }
+
+    const customVarKeys = Object.keys(serverConfig.customUserVars);
+    if (customVarKeys.length === 0) {
+      plugin.authConfig = [];
+    } else {
+      plugin.authConfig = Object.entries(serverConfig.customUserVars).map(([key, value]) => ({
+        authField: key,
+        label: value.title || key,
+        description: value.description || '',
+      }));
+    }
+
+    plugins.push(plugin);
+  }
+
+  return plugins;
+}
+
 module.exports = {
   getAvailableTools,
   getAvailablePluginsController,

api/server/controllers/PluginController.spec.js

@@ -0,0 +1,89 @@
+const { Constants } = require('librechat-data-provider');
+const { getCustomConfig, getCachedTools } = require('~/server/services/Config');
+const { getLogStores } = require('~/cache');
+
+// Mock the dependencies
+jest.mock('@librechat/data-schemas', () => ({
+  logger: {
+    debug: jest.fn(),
+    error: jest.fn(),
+  },
+}));
+
+jest.mock('~/server/services/Config', () => ({
+  getCustomConfig: jest.fn(),
+  getCachedTools: jest.fn(),
+}));
+
+jest.mock('~/server/services/ToolService', () => ({
+  getToolkitKey: jest.fn(),
+}));
+
+jest.mock('~/config', () => ({
+  getMCPManager: jest.fn(() => ({
+    loadManifestTools: jest.fn().mockResolvedValue([]),
+  })),
+  getFlowStateManager: jest.fn(),
+}));
+
+jest.mock('~/app/clients/tools', () => ({
+  availableTools: [],
+}));
+
+jest.mock('~/cache', () => ({
+  getLogStores: jest.fn(),
+}));
+
+// Import the actual module with the function we want to test
+const { getAvailableTools } = require('./PluginController');
+
+describe('PluginController', () => {
+  describe('plugin.icon behavior', () => {
+    let mockReq, mockRes, mockCache;
+
+    const callGetAvailableToolsWithMCPServer = async (mcpServers) => {
+      mockCache.get.mockResolvedValue(null);
+      getCustomConfig.mockResolvedValue({ mcpServers });
+
+      const functionTools = {
+        [`test-tool${Constants.mcp_delimiter}test-server`]: {
+          function: { name: 'test-tool', description: 'A test tool' },
+        },
+      };
+      getCachedTools.mockResolvedValueOnce(functionTools);
+      getCachedTools.mockResolvedValueOnce({
+        [`test-tool${Constants.mcp_delimiter}test-server`]: true,
+      });
+
+      await getAvailableTools(mockReq, mockRes);
+      const responseData = mockRes.json.mock.calls[0][0];
+      return responseData.find((tool) => tool.name === 'test-tool');
+    };
+
+    beforeEach(() => {
+      jest.clearAllMocks();
+      mockReq = { user: { id: 'test-user-id' } };
+      mockRes = { status: jest.fn().mockReturnThis(), json: jest.fn() };
+      mockCache = { get: jest.fn(), set: jest.fn() };
+      getLogStores.mockReturnValue(mockCache);
+    });
+
+    it('should set plugin.icon when iconPath is defined', async () => {
+      const mcpServers = {
+        'test-server': {
+          iconPath: '/path/to/icon.png',
+        },
+      };
+      const testTool = await callGetAvailableToolsWithMCPServer(mcpServers);
+      expect(testTool.icon).toBe('/path/to/icon.png');
+    });
+
+    it('should set plugin.icon to undefined when iconPath is not defined', async () => {
+      const mcpServers = {
+        'test-server': {},
+      };
+      const testTool = await callGetAvailableToolsWithMCPServer(mcpServers);
+      expect(testTool.icon).toBeUndefined();
+    });
+  });
+});

api/server/controllers/UserController.js

@@ -1,11 +1,5 @@
-const {
-  Tools,
-  Constants,
-  FileSources,
-  webSearchKeys,
-  extractWebSearchEnvVars,
-} = require('librechat-data-provider');
 const { logger } = require('@librechat/data-schemas');
+const { webSearchKeys, extractWebSearchEnvVars } = require('@librechat/api');
 const {
   getFiles,
   updateUser,
@@ -20,6 +14,7 @@ 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 { deleteToolCalls } = require('~/models/ToolCall');
@@ -180,14 +175,16 @@ const updateUserPluginsController = async (req, res) => {
         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] Disconnecting MCP connections for user ${user.id} after plugin auth update for ${pluginKey}.`,
+              `[updateUserPluginsController] Disconnecting MCP server ${serverName} for user ${user.id} after plugin auth update for ${pluginKey}.`,
             );
-            await mcpManager.disconnectUserConnections(user.id);
+            await mcpManager.disconnectUserConnection(user.id, serverName);
           }
         } catch (disconnectError) {
           logger.error(
-            `[updateUserPluginsController] Error disconnecting MCP connections for user ${user.id} after plugin auth update:`,
+            `[updateUserPluginsController] Error disconnecting MCP connection for user ${user.id} after plugin auth update:`,
             disconnectError,
           );
           // Do not fail the request for this, but log it.

api/server/controllers/agents/__tests__/v1.spec.js

@@ -0,0 +1,195 @@
+const { duplicateAgent } = require('../v1');
+const { getAgent, createAgent } = require('~/models/Agent');
+const { getActions } = require('~/models/Action');
+const { nanoid } = require('nanoid');
+
+jest.mock('~/models/Agent');
+jest.mock('~/models/Action');
+jest.mock('nanoid');
+
+describe('duplicateAgent', () => {
+  let req, res;
+
+  beforeEach(() => {
+    req = {
+      params: { id: 'agent_123' },
+      user: { id: 'user_456' },
+    };
+    res = {
+      status: jest.fn().mockReturnThis(),
+      json: jest.fn(),
+    };
+    jest.clearAllMocks();
+  });
+
+  it('should duplicate an agent successfully', async () => {
+    const mockAgent = {
+      id: 'agent_123',
+      name: 'Test Agent',
+      description: 'Test Description',
+      instructions: 'Test Instructions',
+      provider: 'openai',
+      model: 'gpt-4',
+      tools: ['file_search'],
+      actions: [],
+      author: 'user_789',
+      versions: [{ name: 'Test Agent', version: 1 }],
+      __v: 0,
+    };
+
+    const mockNewAgent = {
+      id: 'agent_new_123',
+      name: 'Test Agent (1/2/23, 12:34)',
+      description: 'Test Description',
+      instructions: 'Test Instructions',
+      provider: 'openai',
+      model: 'gpt-4',
+      tools: ['file_search'],
+      actions: [],
+      author: 'user_456',
+      versions: [
+        {
+          name: 'Test Agent (1/2/23, 12:34)',
+          description: 'Test Description',
+          instructions: 'Test Instructions',
+          provider: 'openai',
+          model: 'gpt-4',
+          tools: ['file_search'],
+          actions: [],
+          createdAt: new Date(),
+          updatedAt: new Date(),
+        },
+      ],
+    };
+
+    getAgent.mockResolvedValue(mockAgent);
+    getActions.mockResolvedValue([]);
+    nanoid.mockReturnValue('new_123');
+    createAgent.mockResolvedValue(mockNewAgent);
+
+    await duplicateAgent(req, res);
+
+    expect(getAgent).toHaveBeenCalledWith({ id: 'agent_123' });
+    expect(getActions).toHaveBeenCalledWith({ agent_id: 'agent_123' }, true);
+    expect(createAgent).toHaveBeenCalledWith(
+      expect.objectContaining({
+        id: 'agent_new_123',
+        author: 'user_456',
+        name: expect.stringContaining('Test Agent ('),
+        description: 'Test Description',
+        instructions: 'Test Instructions',
+        provider: 'openai',
+        model: 'gpt-4',
+        tools: ['file_search'],
+        actions: [],
+      }),
+    );
+
+    expect(createAgent).toHaveBeenCalledWith(
+      expect.not.objectContaining({
+        versions: expect.anything(),
+        __v: expect.anything(),
+      }),
+    );
+
+    expect(res.status).toHaveBeenCalledWith(201);
+    expect(res.json).toHaveBeenCalledWith({
+      agent: mockNewAgent,
+      actions: [],
+    });
+  });
+
+  it('should ensure duplicated agent has clean versions array without nested fields', async () => {
+    const mockAgent = {
+      id: 'agent_123',
+      name: 'Test Agent',
+      description: 'Test Description',
+      versions: [
+        {
+          name: 'Test Agent',
+          versions: [{ name: 'Nested' }],
+          __v: 1,
+        },
+      ],
+      __v: 2,
+    };
+
+    const mockNewAgent = {
+      id: 'agent_new_123',
+      name: 'Test Agent (1/2/23, 12:34)',
+      description: 'Test Description',
+      versions: [
+        {
+          name: 'Test Agent (1/2/23, 12:34)',
+          description: 'Test Description',
+          createdAt: new Date(),
+          updatedAt: new Date(),
+        },
+      ],
+    };
+
+    getAgent.mockResolvedValue(mockAgent);
+    getActions.mockResolvedValue([]);
+    nanoid.mockReturnValue('new_123');
+    createAgent.mockResolvedValue(mockNewAgent);
+
+    await duplicateAgent(req, res);
+
+    expect(mockNewAgent.versions).toHaveLength(1);
+
+    const firstVersion = mockNewAgent.versions[0];
+    expect(firstVersion).not.toHaveProperty('versions');
+    expect(firstVersion).not.toHaveProperty('__v');
+
+    expect(mockNewAgent).not.toHaveProperty('__v');
+
+    expect(res.status).toHaveBeenCalledWith(201);
+  });
+
+  it('should return 404 if agent not found', async () => {
+    getAgent.mockResolvedValue(null);
+
+    await duplicateAgent(req, res);
+
+    expect(res.status).toHaveBeenCalledWith(404);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Agent not found',
+      status: 'error',
+    });
+  });
+
+  it('should handle tool_resources.ocr correctly', async () => {
+    const mockAgent = {
+      id: 'agent_123',
+      name: 'Test Agent',
+      tool_resources: {
+        ocr: { enabled: true, config: 'test' },
+        other: { should: 'not be copied' },
+      },
+    };
+
+    getAgent.mockResolvedValue(mockAgent);
+    getActions.mockResolvedValue([]);
+    nanoid.mockReturnValue('new_123');
+    createAgent.mockResolvedValue({ id: 'agent_new_123' });
+
+    await duplicateAgent(req, res);
+
+    expect(createAgent).toHaveBeenCalledWith(
+      expect.objectContaining({
+        tool_resources: {
+          ocr: { enabled: true, config: 'test' },
+        },
+      }),
+    );
+  });
+
+  it('should handle errors gracefully', async () => {
+    getAgent.mockRejectedValue(new Error('Database error'));
+
+    await duplicateAgent(req, res);
+
+    expect(res.status).toHaveBeenCalledWith(500);
+    expect(res.json).toHaveBeenCalledWith({ error: 'Database error' });
+  });
+});

api/server/controllers/agents/callbacks.js

@@ -11,6 +11,7 @@ const {
   handleToolCalls,
   ChatModelStreamHandler,
 } = require('@librechat/agents');
+const { processFileCitations } = require('~/server/services/Files/Citations');
 const { processCodeOutput } = require('~/server/services/Files/Code/process');
 const { loadAuthValues } = require('~/server/services/Tools/credentials');
 const { saveBase64Image } = require('~/server/services/Files/process');
@@ -238,6 +239,31 @@ function createToolEndCallback({ req, res, artifactPromises }) {
       return;
     }
 
+    if (output.artifact[Tools.file_search]) {
+      artifactPromises.push(
+        (async () => {
+          const user = req.user;
+          const attachment = await processFileCitations({
+            user,
+            metadata,
+            toolArtifact: output.artifact,
+            toolCallId: output.tool_call_id,
+          });
+          if (!attachment) {
+            return null;
+          }
+          if (!res.headersSent) {
+            return attachment;
+          }
+          res.write(`event: attachment\ndata: ${JSON.stringify(attachment)}\n\n`);
+          return attachment;
+        })().catch((error) => {
+          logger.error('Error processing file citations:', error);
+          return null;
+        }),
+      );
+    }
+
     if (output.artifact[Tools.web_search]) {
       artifactPromises.push(
         (async () => {

api/server/controllers/agents/client.js

@@ -1,18 +1,23 @@
 require('events').EventEmitter.defaultMaxListeners = 100;
 const { logger } = require('@librechat/data-schemas');
+const { DynamicStructuredTool } = require('@langchain/core/tools');
+const { getBufferString, HumanMessage } = require('@langchain/core/messages');
 const {
   sendEvent,
   createRun,
   Tokenizer,
+  checkAccess,
   memoryInstructions,
+  formatContentStrings,
   createMemoryProcessor,
 } = require('@librechat/api');
 const {
   Callback,
+  Providers,
   GraphEvents,
+  TitleMethod,
   formatMessage,
   formatAgentMessages,
-  formatContentStrings,
   getTokenCountForMessage,
   createMetadataAggregator,
 } = require('@librechat/agents');
@@ -22,31 +27,41 @@ const {
   VisionModes,
   ContentTypes,
   EModelEndpoint,
-  KnownEndpoints,
   PermissionTypes,
   isAgentsEndpoint,
   AgentCapabilities,
   bedrockInputSchema,
   removeNullishValues,
 } = require('librechat-data-provider');
-const { DynamicStructuredTool } = require('@langchain/core/tools');
-const { getBufferString, HumanMessage } = require('@langchain/core/messages');
 const {
-  getCustomEndpointConfig,
-  createGetMCPAuthMap,
-  checkCapability,
-} = require('~/server/services/Config');
+  findPluginAuthsByKeys,
+  getFormattedMemories,
+  deleteMemory,
+  setMemory,
+} = require('~/models');
+const { getMCPAuthMap, checkCapability, hasCustomUserVars } = require('~/server/services/Config');
 const { addCacheControl, createContextHandlers } = require('~/app/clients/prompts');
 const { initializeAgent } = require('~/server/services/Endpoints/agents/agent');
 const { spendTokens, spendStructuredTokens } = require('~/models/spendTokens');
-const { getFormattedMemories, deleteMemory, setMemory } = require('~/models');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
-const initOpenAI = require('~/server/services/Endpoints/openAI/initialize');
-const { checkAccess } = require('~/server/middleware/roles/access');
+const { getProviderConfig } = require('~/server/services/Endpoints');
 const BaseClient = require('~/app/clients/BaseClient');
+const { getRoleByName } = require('~/models/Role');
 const { loadAgent } = require('~/models/Agent');
 const { getMCPManager } = require('~/config');
 
+const omitTitleOptions = new Set([
+  'stream',
+  'thinking',
+  'streaming',
+  'clientOptions',
+  'thinkingConfig',
+  'thinkingBudget',
+  'includeThoughts',
+  'maxOutputTokens',
+  'additionalModelRequestFields',
+]);
+
 /**
  * @param {ServerRequest} req
  * @param {Agent} agent
@@ -56,13 +71,15 @@ const payloadParser = ({ req, agent, endpoint }) => {
   if (isAgentsEndpoint(endpoint)) {
     return { model: undefined };
   } else if (endpoint === EModelEndpoint.bedrock) {
-    return bedrockInputSchema.parse(agent.model_parameters);
+    const parsedValues = bedrockInputSchema.parse(agent.model_parameters);
+    if (parsedValues.thinking == null) {
+      parsedValues.thinking = false;
+    }
+    return parsedValues;
   }
   return req.body.endpointOption.model_parameters;
 };
 
-const legacyContentEndpoints = new Set([KnownEndpoints.groq, KnownEndpoints.deepseek]);
-
 const noSystemModelRegex = [/\b(o1-preview|o1-mini|amazon\.titan-text)\b/gi];
 
 function createTokenCounter(encoding) {
@@ -393,7 +410,12 @@ class AgentClient extends BaseClient {
     if (user.personalization?.memories === false) {
       return;
     }
-    const hasAccess = await checkAccess(user, PermissionTypes.MEMORIES, [Permissions.USE]);
+    const hasAccess = await checkAccess({
+      user,
+      permissionType: PermissionTypes.MEMORIES,
+      permissions: [Permissions.USE],
+      getRoleByName,
+    });
 
     if (!hasAccess) {
       logger.debug(
@@ -438,6 +460,12 @@ class AgentClient extends BaseClient {
       res: this.options.res,
       agent: prelimAgent,
       allowedProviders,
+      endpointOption: {
+        endpoint:
+          prelimAgent.id !== Constants.EPHEMERAL_AGENT_ID
+            ? EModelEndpoint.agents
+            : memoryConfig.agent?.provider,
+      },
     });
 
     if (!agent) {
@@ -484,6 +512,39 @@ class AgentClient extends BaseClient {
     return withoutKeys;
   }
 
+  /**
+   * Filters out image URLs from message content
+   * @param {BaseMessage} message - The message to filter
+   * @returns {BaseMessage} - A new message with image URLs removed
+   */
+  filterImageUrls(message) {
+    if (!message.content || typeof message.content === 'string') {
+      return message;
+    }
+
+    if (Array.isArray(message.content)) {
+      const filteredContent = message.content.filter(
+        (part) => part.type !== ContentTypes.IMAGE_URL,
+      );
+
+      if (filteredContent.length === 1 && filteredContent[0].type === ContentTypes.TEXT) {
+        const MessageClass = message.constructor;
+        return new MessageClass({
+          content: filteredContent[0].text,
+          additional_kwargs: message.additional_kwargs,
+        });
+      }
+
+      const MessageClass = message.constructor;
+      return new MessageClass({
+        content: filteredContent,
+        additional_kwargs: message.additional_kwargs,
+      });
+    }
+
+    return message;
+  }
+
   /**
    * @param {BaseMessage[]} messages
    * @returns {Promise<void | (TAttachment | null)[]>}
@@ -511,7 +572,11 @@ class AgentClient extends BaseClient {
           messagesToProcess = [...messages.slice(-messageWindowSize)];
         }
       }
-      return await this.processMemory(messagesToProcess);
+
+      const filteredMessages = messagesToProcess.map((msg) => this.filterImageUrls(msg));
+      const bufferString = getBufferString(filteredMessages);
+      const bufferMessage = new HumanMessage(`# Current Chat:\n\n${bufferString}`);
+      return await this.processMemory([bufferMessage]);
     } catch (error) {
       logger.error('Memory Agent failed to process memory', error);
     }
@@ -677,23 +742,18 @@ class AgentClient extends BaseClient {
           hide_sequential_outputs: this.options.agent.hide_sequential_outputs,
           user: this.options.req.user,
         },
-        recursionLimit: agentsEConfig?.recursionLimit,
+        recursionLimit: agentsEConfig?.recursionLimit ?? 25,
         signal: abortController.signal,
         streamMode: 'values',
         version: 'v2',
       };
 
-      const getUserMCPAuthMap = await createGetMCPAuthMap();
-
       const toolSet = new Set((this.options.agent.tools ?? []).map((tool) => tool && tool.name));
       let { messages: initialMessages, indexTokenCountMap } = formatAgentMessages(
         payload,
         this.indexTokenCountMap,
         toolSet,
       );
-      if (legacyContentEndpoints.has(this.options.agent.endpoint?.toLowerCase())) {
-        initialMessages = formatContentStrings(initialMessages);
-      }
 
       /**
        *
@@ -709,6 +769,9 @@ class AgentClient extends BaseClient {
         if (i > 0) {
           this.model = agent.model_parameters.model;
         }
+        if (i > 0 && config.signal == null) {
+          config.signal = abortController.signal;
+        }
         if (agent.recursion_limit && typeof agent.recursion_limit === 'number') {
           config.recursionLimit = agent.recursion_limit;
         }
@@ -747,7 +810,7 @@ class AgentClient extends BaseClient {
 
         if (noSystemMessages === true && systemContent?.length) {
           const latestMessageContent = _messages.pop().content;
-          if (typeof latestMessage !== 'string') {
+          if (typeof latestMessageContent !== 'string') {
             latestMessageContent[0].text = [systemContent, latestMessageContent[0].text].join('\n');
             _messages.push(new HumanMessage({ content: latestMessageContent }));
           } else {
@@ -757,6 +820,9 @@ class AgentClient extends BaseClient {
         }
 
         let messages = _messages;
+        if (agent.useLegacyContent === true) {
+          messages = formatContentStrings(messages);
+        }
         if (
           agent.model_parameters?.clientOptions?.defaultHeaders?.['anthropic-beta']?.includes(
             'prompt-caching',
@@ -805,10 +871,11 @@ class AgentClient extends BaseClient {
         }
 
         try {
-          if (getUserMCPAuthMap) {
-            config.configurable.userMCPAuthMap = await getUserMCPAuthMap({
+          if (await hasCustomUserVars()) {
+            config.configurable.userMCPAuthMap = await getMCPAuthMap({
               tools: agent.tools,
               userId: this.options.req.user.id,
+              findPluginAuthsByKeys,
             });
           }
         } catch (err) {
@@ -941,6 +1008,7 @@ class AgentClient extends BaseClient {
             this.artifactPromises.push(...attachments);
           }
         }
+
         await this.recordCollectedUsage({ context: 'message' });
       } catch (err) {
         logger.error(
@@ -983,23 +1051,41 @@ class AgentClient extends BaseClient {
       throw new Error('Run not initialized');
     }
     const { handleLLMEnd, collected: collectedMetadata } = createMetadataAggregator();
-    const endpoint = this.options.agent.endpoint;
-    const { req, res } = this.options;
+    const { req, res, agent } = this.options;
+    let endpoint = agent.endpoint;
+
     /** @type {import('@librechat/agents').ClientOptions} */
     let clientOptions = {
       maxTokens: 75,
+      model: agent.model || agent.model_parameters.model,
     };
-    let endpointConfig = req.app.locals[endpoint];
+
+    let titleProviderConfig = await getProviderConfig(endpoint);
+
+    /** @type {TEndpoint | undefined} */
+    const endpointConfig =
+      req.app.locals.all ?? req.app.locals[endpoint] ?? titleProviderConfig.customEndpointConfig;
     if (!endpointConfig) {
+      logger.warn(
+        '[api/server/controllers/agents/client.js #titleConvo] Error getting endpoint config',
+      );
+    }
+
+    if (endpointConfig?.titleEndpoint && endpointConfig.titleEndpoint !== endpoint) {
       try {
-        endpointConfig = await getCustomEndpointConfig(endpoint);
-      } catch (err) {
-        logger.error(
-          '[api/server/controllers/agents/client.js #titleConvo] Error getting custom endpoint config',
-          err,
+        titleProviderConfig = await getProviderConfig(endpointConfig.titleEndpoint);
+        endpoint = endpointConfig.titleEndpoint;
+      } catch (error) {
+        logger.warn(
+          `[api/server/controllers/agents/client.js #titleConvo] Error getting title endpoint config for ${endpointConfig.titleEndpoint}, falling back to default`,
+          error,
         );
+        // Fall back to original provider config
+        endpoint = agent.endpoint;
+        titleProviderConfig = await getProviderConfig(endpoint);
       }
     }
+
     if (
       endpointConfig &&
       endpointConfig.titleModel &&
@@ -1007,33 +1093,66 @@ class AgentClient extends BaseClient {
     ) {
       clientOptions.model = endpointConfig.titleModel;
     }
+
+    const options = await titleProviderConfig.getOptions({
+      req,
+      res,
+      optionsOnly: true,
+      overrideEndpoint: endpoint,
+      overrideModel: clientOptions.model,
+      endpointOption: { model_parameters: clientOptions },
+    });
+
+    let provider = options.provider ?? titleProviderConfig.overrideProvider ?? agent.provider;
     if (
       endpoint === EModelEndpoint.azureOpenAI &&
-      clientOptions.model &&
-      this.options.agent.model_parameters.model !== clientOptions.model
+      options.llmConfig?.azureOpenAIApiInstanceName == null
     ) {
-      clientOptions =
-        (
-          await initOpenAI({
-            req,
-            res,
-            optionsOnly: true,
-            overrideModel: clientOptions.model,
-            overrideEndpoint: endpoint,
-            endpointOption: {
-              model_parameters: clientOptions,
-            },
-          })
-        )?.llmConfig ?? clientOptions;
+      provider = Providers.OPENAI;
+    } else if (
+      endpoint === EModelEndpoint.azureOpenAI &&
+      options.llmConfig?.azureOpenAIApiInstanceName != null &&
+      provider !== Providers.AZURE
+    ) {
+      provider = Providers.AZURE;
     }
-    if (/\b(o\d)\b/i.test(clientOptions.model) && clientOptions.maxTokens != null) {
+
+    /** @type {import('@librechat/agents').ClientOptions} */
+    clientOptions = { ...options.llmConfig };
+    if (options.configOptions) {
+      clientOptions.configuration = options.configOptions;
+    }
+
+    // Ensure maxTokens is set for non-o1 models
+    if (!/\b(o\d)\b/i.test(clientOptions.model) && !clientOptions.maxTokens) {
+      clientOptions.maxTokens = 75;
+    } else if (/\b(o\d)\b/i.test(clientOptions.model) && clientOptions.maxTokens != null) {
       delete clientOptions.maxTokens;
     }
+
+    clientOptions = Object.assign(
+      Object.fromEntries(
+        Object.entries(clientOptions).filter(([key]) => !omitTitleOptions.has(key)),
+      ),
+    );
+
+    if (
+      provider === Providers.GOOGLE &&
+      (endpointConfig?.titleMethod === TitleMethod.FUNCTIONS ||
+        endpointConfig?.titleMethod === TitleMethod.STRUCTURED)
+    ) {
+      clientOptions.json = true;
+    }
+
     try {
       const titleResult = await this.run.generateTitle({
+        provider,
+        clientOptions,
         inputText: text,
         contentParts: this.contentParts,
-        clientOptions,
+        titleMethod: endpointConfig?.titleMethod,
+        titlePrompt: endpointConfig?.titlePrompt,
+        titlePromptTemplate: endpointConfig?.titlePromptTemplate,
         chainOptions: {
           signal: abortController.signal,
           callbacks: [
@@ -1048,8 +1167,10 @@ class AgentClient extends BaseClient {
         let input_tokens, output_tokens;
 
         if (item.usage) {
-          input_tokens = item.usage.input_tokens || item.usage.inputTokens;
-          output_tokens = item.usage.output_tokens || item.usage.outputTokens;
+          input_tokens =
+            item.usage.prompt_tokens || item.usage.input_tokens || item.usage.inputTokens;
+          output_tokens =
+            item.usage.completion_tokens || item.usage.output_tokens || item.usage.outputTokens;
         } else if (item.tokenUsage) {
           input_tokens = item.tokenUsage.promptTokens;
           output_tokens = item.tokenUsage.completionTokens;
@@ -1079,8 +1200,52 @@ class AgentClient extends BaseClient {
     }
   }
 
-  /** Silent method, as `recordCollectedUsage` is used instead */
-  async recordTokenUsage() {}
+  /**
+   * @param {object} params
+   * @param {number} params.promptTokens
+   * @param {number} params.completionTokens
+   * @param {OpenAIUsageMetadata} [params.usage]
+   * @param {string} [params.model]
+   * @param {string} [params.context='message']
+   * @returns {Promise<void>}
+   */
+  async recordTokenUsage({ model, promptTokens, completionTokens, usage, context = 'message' }) {
+    try {
+      await spendTokens(
+        {
+          model,
+          context,
+          conversationId: this.conversationId,
+          user: this.user ?? this.options.req.user?.id,
+          endpointTokenConfig: this.options.endpointTokenConfig,
+        },
+        { promptTokens, completionTokens },
+      );
+
+      if (
+        usage &&
+        typeof usage === 'object' &&
+        'reasoning_tokens' in usage &&
+        typeof usage.reasoning_tokens === 'number'
+      ) {
+        await spendTokens(
+          {
+            model,
+            context: 'reasoning',
+            conversationId: this.conversationId,
+            user: this.user ?? this.options.req.user?.id,
+            endpointTokenConfig: this.options.endpointTokenConfig,
+          },
+          { completionTokens: usage.reasoning_tokens },
+        );
+      }
+    } catch (error) {
+      logger.error(
+        '[api/server/controllers/agents/client.js #recordTokenUsage] Error recording token usage',
+        error,
+      );
+    }
+  }
 
   getEncoding() {
     return 'o200k_base';

api/server/controllers/agents/client.test.js

@@ -0,0 +1,957 @@
+const { Providers } = require('@librechat/agents');
+const { Constants, EModelEndpoint } = require('librechat-data-provider');
+const AgentClient = require('./client');
+
+jest.mock('@librechat/agents', () => ({
+  ...jest.requireActual('@librechat/agents'),
+  createMetadataAggregator: () => ({
+    handleLLMEnd: jest.fn(),
+    collected: [],
+  }),
+}));
+
+describe('AgentClient - titleConvo', () => {
+  let client;
+  let mockRun;
+  let mockReq;
+  let mockRes;
+  let mockAgent;
+  let mockOptions;
+
+  beforeEach(() => {
+    // Reset all mocks
+    jest.clearAllMocks();
+
+    // Mock run object
+    mockRun = {
+      generateTitle: jest.fn().mockResolvedValue({
+        title: 'Generated Title',
+      }),
+    };
+
+    // Mock agent - with both endpoint and provider
+    mockAgent = {
+      id: 'agent-123',
+      endpoint: EModelEndpoint.openAI, // Use a valid provider as endpoint for getProviderConfig
+      provider: EModelEndpoint.openAI, // Add provider property
+      model_parameters: {
+        model: 'gpt-4',
+      },
+    };
+
+    // Mock request and response
+    mockReq = {
+      app: {
+        locals: {
+          [EModelEndpoint.openAI]: {
+            // Match the agent endpoint
+            titleModel: 'gpt-3.5-turbo',
+            titlePrompt: 'Custom title prompt',
+            titleMethod: 'structured',
+            titlePromptTemplate: 'Template: {{content}}',
+          },
+        },
+      },
+      user: {
+        id: 'user-123',
+      },
+      body: {
+        model: 'gpt-4',
+        endpoint: EModelEndpoint.openAI,
+        key: null,
+      },
+    };
+
+    mockRes = {};
+
+    // Mock options
+    mockOptions = {
+      req: mockReq,
+      res: mockRes,
+      agent: mockAgent,
+      endpointTokenConfig: {},
+    };
+
+    // Create client instance
+    client = new AgentClient(mockOptions);
+    client.run = mockRun;
+    client.responseMessageId = 'response-123';
+    client.conversationId = 'convo-123';
+    client.contentParts = [{ type: 'text', text: 'Test content' }];
+    client.recordCollectedUsage = jest.fn().mockResolvedValue(); // Mock as async function that resolves
+  });
+
+  describe('titleConvo method', () => {
+    it('should throw error if run is not initialized', async () => {
+      client.run = null;
+
+      await expect(
+        client.titleConvo({ text: 'Test', abortController: new AbortController() }),
+      ).rejects.toThrow('Run not initialized');
+    });
+
+    it('should use titlePrompt from endpoint config', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titlePrompt: 'Custom title prompt',
+        }),
+      );
+    });
+
+    it('should use titlePromptTemplate from endpoint config', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titlePromptTemplate: 'Template: {{content}}',
+        }),
+      );
+    });
+
+    it('should use titleMethod from endpoint config', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          provider: Providers.OPENAI,
+          titleMethod: 'structured',
+        }),
+      );
+    });
+
+    it('should use titleModel from endpoint config when provided', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      // Check that generateTitle was called with correct clientOptions
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('gpt-3.5-turbo');
+    });
+
+    it('should handle missing endpoint config gracefully', async () => {
+      // Remove endpoint config
+      mockReq.app.locals[EModelEndpoint.openAI] = undefined;
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titlePrompt: undefined,
+          titlePromptTemplate: undefined,
+          titleMethod: undefined,
+        }),
+      );
+    });
+
+    it('should use agent model when titleModel is not provided', async () => {
+      // Remove titleModel from config
+      delete mockReq.app.locals[EModelEndpoint.openAI].titleModel;
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('gpt-4'); // Should use agent's model
+    });
+
+    it('should not use titleModel when it equals CURRENT_MODEL constant', async () => {
+      mockReq.app.locals[EModelEndpoint.openAI].titleModel = Constants.CURRENT_MODEL;
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('gpt-4'); // Should use agent's model
+    });
+
+    it('should pass all required parameters to generateTitle', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(mockRun.generateTitle).toHaveBeenCalledWith({
+        provider: expect.any(String),
+        inputText: text,
+        contentParts: client.contentParts,
+        clientOptions: expect.objectContaining({
+          model: 'gpt-3.5-turbo',
+        }),
+        titlePrompt: 'Custom title prompt',
+        titlePromptTemplate: 'Template: {{content}}',
+        titleMethod: 'structured',
+        chainOptions: expect.objectContaining({
+          signal: abortController.signal,
+        }),
+      });
+    });
+
+    it('should record collected usage after title generation', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      expect(client.recordCollectedUsage).toHaveBeenCalledWith({
+        model: 'gpt-3.5-turbo',
+        context: 'title',
+        collectedUsage: expect.any(Array),
+      });
+    });
+
+    it('should return the generated title', async () => {
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      const result = await client.titleConvo({ text, abortController });
+
+      expect(result).toBe('Generated Title');
+    });
+
+    it('should handle errors gracefully and return undefined', async () => {
+      mockRun.generateTitle.mockRejectedValue(new Error('Title generation failed'));
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      const result = await client.titleConvo({ text, abortController });
+
+      expect(result).toBeUndefined();
+    });
+
+    it('should pass titleEndpoint configuration to generateTitle', async () => {
+      // Mock the API key just for this test
+      const originalApiKey = process.env.ANTHROPIC_API_KEY;
+      process.env.ANTHROPIC_API_KEY = 'test-api-key';
+
+      // Add titleEndpoint to the config
+      mockReq.app.locals[EModelEndpoint.openAI].titleEndpoint = EModelEndpoint.anthropic;
+      mockReq.app.locals[EModelEndpoint.openAI].titleMethod = 'structured';
+      mockReq.app.locals[EModelEndpoint.openAI].titlePrompt = 'Custom title prompt';
+      mockReq.app.locals[EModelEndpoint.openAI].titlePromptTemplate = 'Custom template';
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      // Verify generateTitle was called with the custom configuration
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titleMethod: 'structured',
+          provider: Providers.ANTHROPIC,
+          titlePrompt: 'Custom title prompt',
+          titlePromptTemplate: 'Custom template',
+        }),
+      );
+
+      // Restore the original API key
+      if (originalApiKey) {
+        process.env.ANTHROPIC_API_KEY = originalApiKey;
+      } else {
+        delete process.env.ANTHROPIC_API_KEY;
+      }
+    });
+
+    it('should use all config when endpoint config is missing', async () => {
+      // Remove endpoint-specific config
+      delete mockReq.app.locals[EModelEndpoint.openAI].titleModel;
+      delete mockReq.app.locals[EModelEndpoint.openAI].titlePrompt;
+      delete mockReq.app.locals[EModelEndpoint.openAI].titleMethod;
+      delete mockReq.app.locals[EModelEndpoint.openAI].titlePromptTemplate;
+
+      // Set 'all' config
+      mockReq.app.locals.all = {
+        titleModel: 'gpt-4o-mini',
+        titlePrompt: 'All config title prompt',
+        titleMethod: 'completion',
+        titlePromptTemplate: 'All config template: {{content}}',
+      };
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      // Verify generateTitle was called with 'all' config values
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titleMethod: 'completion',
+          titlePrompt: 'All config title prompt',
+          titlePromptTemplate: 'All config template: {{content}}',
+        }),
+      );
+
+      // Check that the model was set from 'all' config
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('gpt-4o-mini');
+    });
+
+    it('should prioritize all config over endpoint config for title settings', async () => {
+      // Set both endpoint and 'all' config
+      mockReq.app.locals[EModelEndpoint.openAI].titleModel = 'gpt-3.5-turbo';
+      mockReq.app.locals[EModelEndpoint.openAI].titlePrompt = 'Endpoint title prompt';
+      mockReq.app.locals[EModelEndpoint.openAI].titleMethod = 'structured';
+      // Remove titlePromptTemplate from endpoint config to test fallback
+      delete mockReq.app.locals[EModelEndpoint.openAI].titlePromptTemplate;
+
+      mockReq.app.locals.all = {
+        titleModel: 'gpt-4o-mini',
+        titlePrompt: 'All config title prompt',
+        titleMethod: 'completion',
+        titlePromptTemplate: 'All config template',
+      };
+
+      const text = 'Test conversation text';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      // Verify 'all' config takes precedence over endpoint config
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          titleMethod: 'completion',
+          titlePrompt: 'All config title prompt',
+          titlePromptTemplate: 'All config template',
+        }),
+      );
+
+      // Check that the model was set from 'all' config
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('gpt-4o-mini');
+    });
+
+    it('should use all config with titleEndpoint and verify provider switch', async () => {
+      // Mock the API key for the titleEndpoint provider
+      const originalApiKey = process.env.ANTHROPIC_API_KEY;
+      process.env.ANTHROPIC_API_KEY = 'test-anthropic-key';
+
+      // Remove endpoint-specific config to test 'all' config
+      delete mockReq.app.locals[EModelEndpoint.openAI];
+
+      // Set comprehensive 'all' config with all new title options
+      mockReq.app.locals.all = {
+        titleConvo: true,
+        titleModel: 'claude-3-haiku-20240307',
+        titleMethod: 'completion', // Testing the new default method
+        titlePrompt: 'Generate a concise, descriptive title for this conversation',
+        titlePromptTemplate: 'Conversation summary: {{content}}',
+        titleEndpoint: EModelEndpoint.anthropic, // Should switch provider to Anthropic
+      };
+
+      const text = 'Test conversation about AI and machine learning';
+      const abortController = new AbortController();
+
+      await client.titleConvo({ text, abortController });
+
+      // Verify all config values were used
+      expect(mockRun.generateTitle).toHaveBeenCalledWith(
+        expect.objectContaining({
+          provider: Providers.ANTHROPIC, // Critical: Verify provider switched to Anthropic
+          titleMethod: 'completion',
+          titlePrompt: 'Generate a concise, descriptive title for this conversation',
+          titlePromptTemplate: 'Conversation summary: {{content}}',
+          inputText: text,
+          contentParts: client.contentParts,
+        }),
+      );
+
+      // Verify the model was set from 'all' config
+      const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+      expect(generateTitleCall.clientOptions.model).toBe('claude-3-haiku-20240307');
+
+      // Verify other client options are set correctly
+      expect(generateTitleCall.clientOptions).toMatchObject({
+        model: 'claude-3-haiku-20240307',
+        // Note: Anthropic's getOptions may set its own maxTokens value
+      });
+
+      // Restore the original API key
+      if (originalApiKey) {
+        process.env.ANTHROPIC_API_KEY = originalApiKey;
+      } else {
+        delete process.env.ANTHROPIC_API_KEY;
+      }
+    });
+
+    it('should test all titleMethod options from all config', async () => {
+      // Test each titleMethod: 'completion', 'functions', 'structured'
+      const titleMethods = ['completion', 'functions', 'structured'];
+
+      for (const method of titleMethods) {
+        // Clear previous calls
+        mockRun.generateTitle.mockClear();
+
+        // Remove endpoint config
+        delete mockReq.app.locals[EModelEndpoint.openAI];
+
+        // Set 'all' config with specific titleMethod
+        mockReq.app.locals.all = {
+          titleModel: 'gpt-4o-mini',
+          titleMethod: method,
+          titlePrompt: `Testing ${method} method`,
+          titlePromptTemplate: `Template for ${method}: {{content}}`,
+        };
+
+        const text = `Test conversation for ${method} method`;
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify the correct titleMethod was used
+        expect(mockRun.generateTitle).toHaveBeenCalledWith(
+          expect.objectContaining({
+            titleMethod: method,
+            titlePrompt: `Testing ${method} method`,
+            titlePromptTemplate: `Template for ${method}: {{content}}`,
+          }),
+        );
+      }
+    });
+
+    describe('Azure-specific title generation', () => {
+      let originalEnv;
+
+      beforeEach(() => {
+        // Reset mocks
+        jest.clearAllMocks();
+
+        // Save original environment variables
+        originalEnv = { ...process.env };
+
+        // Mock Azure API keys
+        process.env.AZURE_OPENAI_API_KEY = 'test-azure-key';
+        process.env.AZURE_API_KEY = 'test-azure-key';
+        process.env.EASTUS_API_KEY = 'test-eastus-key';
+        process.env.EASTUS2_API_KEY = 'test-eastus2-key';
+      });
+
+      afterEach(() => {
+        // Restore environment variables
+        process.env = originalEnv;
+      });
+
+      it('should use OPENAI provider for Azure serverless endpoints', async () => {
+        // Set up Azure endpoint with serverless config
+        mockAgent.endpoint = EModelEndpoint.azureOpenAI;
+        mockAgent.provider = EModelEndpoint.azureOpenAI;
+        mockReq.app.locals[EModelEndpoint.azureOpenAI] = {
+          titleConvo: true,
+          titleModel: 'grok-3',
+          titleMethod: 'completion',
+          titlePrompt: 'Azure serverless title prompt',
+          streamRate: 35,
+          modelGroupMap: {
+            'grok-3': {
+              group: 'Azure AI Foundry',
+              deploymentName: 'grok-3',
+            },
+          },
+          groupMap: {
+            'Azure AI Foundry': {
+              apiKey: '${AZURE_API_KEY}',
+              baseURL: 'https://test.services.ai.azure.com/models',
+              version: '2024-05-01-preview',
+              serverless: true,
+              models: {
+                'grok-3': {
+                  deploymentName: 'grok-3',
+                },
+              },
+            },
+          },
+        };
+        mockReq.body.endpoint = EModelEndpoint.azureOpenAI;
+        mockReq.body.model = 'grok-3';
+
+        const text = 'Test Azure serverless conversation';
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify provider was switched to OPENAI for serverless
+        expect(mockRun.generateTitle).toHaveBeenCalledWith(
+          expect.objectContaining({
+            provider: Providers.OPENAI, // Should be OPENAI for serverless
+            titleMethod: 'completion',
+            titlePrompt: 'Azure serverless title prompt',
+          }),
+        );
+      });
+
+      it('should use AZURE provider for Azure endpoints with instanceName', async () => {
+        // Set up Azure endpoint
+        mockAgent.endpoint = EModelEndpoint.azureOpenAI;
+        mockAgent.provider = EModelEndpoint.azureOpenAI;
+        mockReq.app.locals[EModelEndpoint.azureOpenAI] = {
+          titleConvo: true,
+          titleModel: 'gpt-4o',
+          titleMethod: 'structured',
+          titlePrompt: 'Azure instance title prompt',
+          streamRate: 35,
+          modelGroupMap: {
+            'gpt-4o': {
+              group: 'eastus',
+              deploymentName: 'gpt-4o',
+            },
+          },
+          groupMap: {
+            eastus: {
+              apiKey: '${EASTUS_API_KEY}',
+              instanceName: 'region-instance',
+              version: '2024-02-15-preview',
+              models: {
+                'gpt-4o': {
+                  deploymentName: 'gpt-4o',
+                },
+              },
+            },
+          },
+        };
+        mockReq.body.endpoint = EModelEndpoint.azureOpenAI;
+        mockReq.body.model = 'gpt-4o';
+
+        const text = 'Test Azure instance conversation';
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify provider remains AZURE with instanceName
+        expect(mockRun.generateTitle).toHaveBeenCalledWith(
+          expect.objectContaining({
+            provider: Providers.AZURE,
+            titleMethod: 'structured',
+            titlePrompt: 'Azure instance title prompt',
+          }),
+        );
+      });
+
+      it('should handle Azure titleModel with CURRENT_MODEL constant', async () => {
+        // Set up Azure endpoint
+        mockAgent.endpoint = EModelEndpoint.azureOpenAI;
+        mockAgent.provider = EModelEndpoint.azureOpenAI;
+        mockAgent.model_parameters.model = 'gpt-4o-latest';
+        mockReq.app.locals[EModelEndpoint.azureOpenAI] = {
+          titleConvo: true,
+          titleModel: Constants.CURRENT_MODEL,
+          titleMethod: 'functions',
+          streamRate: 35,
+          modelGroupMap: {
+            'gpt-4o-latest': {
+              group: 'region-eastus',
+              deploymentName: 'gpt-4o-mini',
+              version: '2024-02-15-preview',
+            },
+          },
+          groupMap: {
+            'region-eastus': {
+              apiKey: '${EASTUS2_API_KEY}',
+              instanceName: 'test-instance',
+              version: '2024-12-01-preview',
+              models: {
+                'gpt-4o-latest': {
+                  deploymentName: 'gpt-4o-mini',
+                  version: '2024-02-15-preview',
+                },
+              },
+            },
+          },
+        };
+        mockReq.body.endpoint = EModelEndpoint.azureOpenAI;
+        mockReq.body.model = 'gpt-4o-latest';
+
+        const text = 'Test Azure current model';
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify it uses the correct model when titleModel is CURRENT_MODEL
+        const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+        // When CURRENT_MODEL is used with Azure, the model gets mapped to the deployment name
+        // In this case, 'gpt-4o-latest' is mapped to 'gpt-4o-mini' deployment
+        expect(generateTitleCall.clientOptions.model).toBe('gpt-4o-mini');
+        // Also verify that CURRENT_MODEL constant was not passed as the model
+        expect(generateTitleCall.clientOptions.model).not.toBe(Constants.CURRENT_MODEL);
+      });
+
+      it('should handle Azure with multiple model groups', async () => {
+        // Set up Azure endpoint
+        mockAgent.endpoint = EModelEndpoint.azureOpenAI;
+        mockAgent.provider = EModelEndpoint.azureOpenAI;
+        mockReq.app.locals[EModelEndpoint.azureOpenAI] = {
+          titleConvo: true,
+          titleModel: 'o1-mini',
+          titleMethod: 'completion',
+          streamRate: 35,
+          modelGroupMap: {
+            'gpt-4o': {
+              group: 'eastus',
+              deploymentName: 'gpt-4o',
+            },
+            'o1-mini': {
+              group: 'region-eastus',
+              deploymentName: 'o1-mini',
+            },
+            'codex-mini': {
+              group: 'codex-mini',
+              deploymentName: 'codex-mini',
+            },
+          },
+          groupMap: {
+            eastus: {
+              apiKey: '${EASTUS_API_KEY}',
+              instanceName: 'region-eastus',
+              version: '2024-02-15-preview',
+              models: {
+                'gpt-4o': {
+                  deploymentName: 'gpt-4o',
+                },
+              },
+            },
+            'region-eastus': {
+              apiKey: '${EASTUS2_API_KEY}',
+              instanceName: 'region-eastus2',
+              version: '2024-12-01-preview',
+              models: {
+                'o1-mini': {
+                  deploymentName: 'o1-mini',
+                },
+              },
+            },
+            'codex-mini': {
+              apiKey: '${AZURE_API_KEY}',
+              baseURL: 'https://example.cognitiveservices.azure.com/openai/',
+              version: '2025-04-01-preview',
+              serverless: true,
+              models: {
+                'codex-mini': {
+                  deploymentName: 'codex-mini',
+                },
+              },
+            },
+          },
+        };
+        mockReq.body.endpoint = EModelEndpoint.azureOpenAI;
+        mockReq.body.model = 'o1-mini';
+
+        const text = 'Test Azure multi-group conversation';
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify correct model and provider are used
+        expect(mockRun.generateTitle).toHaveBeenCalledWith(
+          expect.objectContaining({
+            provider: Providers.AZURE,
+            titleMethod: 'completion',
+          }),
+        );
+
+        const generateTitleCall = mockRun.generateTitle.mock.calls[0][0];
+        expect(generateTitleCall.clientOptions.model).toBe('o1-mini');
+        expect(generateTitleCall.clientOptions.maxTokens).toBeUndefined(); // o1 models shouldn't have maxTokens
+      });
+
+      it('should use all config as fallback for Azure endpoints', async () => {
+        // Set up Azure endpoint with minimal config
+        mockAgent.endpoint = EModelEndpoint.azureOpenAI;
+        mockAgent.provider = EModelEndpoint.azureOpenAI;
+        mockReq.body.endpoint = EModelEndpoint.azureOpenAI;
+        mockReq.body.model = 'gpt-4';
+
+        // Remove Azure-specific config
+        delete mockReq.app.locals[EModelEndpoint.azureOpenAI];
+
+        // Set 'all' config as fallback with a serverless Azure config
+        mockReq.app.locals.all = {
+          titleConvo: true,
+          titleModel: 'gpt-4',
+          titleMethod: 'structured',
+          titlePrompt: 'Fallback title prompt from all config',
+          titlePromptTemplate: 'Template: {{content}}',
+          modelGroupMap: {
+            'gpt-4': {
+              group: 'default-group',
+              deploymentName: 'gpt-4',
+            },
+          },
+          groupMap: {
+            'default-group': {
+              apiKey: '${AZURE_API_KEY}',
+              baseURL: 'https://default.openai.azure.com/',
+              version: '2024-02-15-preview',
+              serverless: true,
+              models: {
+                'gpt-4': {
+                  deploymentName: 'gpt-4',
+                },
+              },
+            },
+          },
+        };
+
+        const text = 'Test Azure with all config fallback';
+        const abortController = new AbortController();
+
+        await client.titleConvo({ text, abortController });
+
+        // Verify all config is used
+        expect(mockRun.generateTitle).toHaveBeenCalledWith(
+          expect.objectContaining({
+            provider: Providers.OPENAI, // Should be OPENAI when no instanceName
+            titleMethod: 'structured',
+            titlePrompt: 'Fallback title prompt from all config',
+            titlePromptTemplate: 'Template: {{content}}',
+          }),
+        );
+      });
+    });
+  });
+
+  describe('runMemory method', () => {
+    let client;
+    let mockReq;
+    let mockRes;
+    let mockAgent;
+    let mockOptions;
+    let mockProcessMemory;
+
+    beforeEach(() => {
+      jest.clearAllMocks();
+
+      mockAgent = {
+        id: 'agent-123',
+        endpoint: EModelEndpoint.openAI,
+        provider: EModelEndpoint.openAI,
+        model_parameters: {
+          model: 'gpt-4',
+        },
+      };
+
+      mockReq = {
+        app: {
+          locals: {
+            memory: {
+              messageWindowSize: 3,
+            },
+          },
+        },
+        user: {
+          id: 'user-123',
+          personalization: {
+            memories: true,
+          },
+        },
+      };
+
+      mockRes = {};
+
+      mockOptions = {
+        req: mockReq,
+        res: mockRes,
+        agent: mockAgent,
+      };
+
+      mockProcessMemory = jest.fn().mockResolvedValue([]);
+
+      client = new AgentClient(mockOptions);
+      client.processMemory = mockProcessMemory;
+      client.conversationId = 'convo-123';
+      client.responseMessageId = 'response-123';
+    });
+
+    it('should filter out image URLs from message content', async () => {
+      const { HumanMessage, AIMessage } = require('@langchain/core/messages');
+      const messages = [
+        new HumanMessage({
+          content: [
+            {
+              type: 'text',
+              text: 'What is in this image?',
+            },
+            {
+              type: 'image_url',
+              image_url: {
+                url: 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNkYPhfDwAChwGA60e6kgAAAABJRU5ErkJggg==',
+                detail: 'auto',
+              },
+            },
+          ],
+        }),
+        new AIMessage('I can see a small red pixel in the image.'),
+        new HumanMessage({
+          content: [
+            {
+              type: 'text',
+              text: 'What about this one?',
+            },
+            {
+              type: 'image_url',
+              image_url: {
+                url: 'data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABgAAD/',
+                detail: 'high',
+              },
+            },
+          ],
+        }),
+      ];
+
+      await client.runMemory(messages);
+
+      expect(mockProcessMemory).toHaveBeenCalledTimes(1);
+      const processedMessage = mockProcessMemory.mock.calls[0][0][0];
+
+      // Verify the buffer message was created
+      expect(processedMessage.constructor.name).toBe('HumanMessage');
+      expect(processedMessage.content).toContain('# Current Chat:');
+
+      // Verify that image URLs are not in the buffer string
+      expect(processedMessage.content).not.toContain('image_url');
+      expect(processedMessage.content).not.toContain('data:image');
+      expect(processedMessage.content).not.toContain('base64');
+
+      // Verify text content is preserved
+      expect(processedMessage.content).toContain('What is in this image?');
+      expect(processedMessage.content).toContain('I can see a small red pixel in the image.');
+      expect(processedMessage.content).toContain('What about this one?');
+    });
+
+    it('should handle messages with only text content', async () => {
+      const { HumanMessage, AIMessage } = require('@langchain/core/messages');
+      const messages = [
+        new HumanMessage('Hello, how are you?'),
+        new AIMessage('I am doing well, thank you!'),
+        new HumanMessage('That is great to hear.'),
+      ];
+
+      await client.runMemory(messages);
+
+      expect(mockProcessMemory).toHaveBeenCalledTimes(1);
+      const processedMessage = mockProcessMemory.mock.calls[0][0][0];
+
+      expect(processedMessage.content).toContain('Hello, how are you?');
+      expect(processedMessage.content).toContain('I am doing well, thank you!');
+      expect(processedMessage.content).toContain('That is great to hear.');
+    });
+
+    it('should handle mixed content types correctly', async () => {
+      const { HumanMessage } = require('@langchain/core/messages');
+      const { ContentTypes } = require('librechat-data-provider');
+
+      const messages = [
+        new HumanMessage({
+          content: [
+            {
+              type: 'text',
+              text: 'Here is some text',
+            },
+            {
+              type: ContentTypes.IMAGE_URL,
+              image_url: {
+                url: 'https://example.com/image.png',
+              },
+            },
+            {
+              type: 'text',
+              text: ' and more text',
+            },
+          ],
+        }),
+      ];
+
+      await client.runMemory(messages);
+
+      expect(mockProcessMemory).toHaveBeenCalledTimes(1);
+      const processedMessage = mockProcessMemory.mock.calls[0][0][0];
+
+      // Should contain text parts but not image URLs
+      expect(processedMessage.content).toContain('Here is some text');
+      expect(processedMessage.content).toContain('and more text');
+      expect(processedMessage.content).not.toContain('example.com/image.png');
+      expect(processedMessage.content).not.toContain('IMAGE_URL');
+    });
+
+    it('should preserve original messages without mutation', async () => {
+      const { HumanMessage } = require('@langchain/core/messages');
+      const originalContent = [
+        {
+          type: 'text',
+          text: 'Original text',
+        },
+        {
+          type: 'image_url',
+          image_url: {
+            url: 'data:image/png;base64,ABC123',
+          },
+        },
+      ];
+
+      const messages = [
+        new HumanMessage({
+          content: [...originalContent],
+        }),
+      ];
+
+      await client.runMemory(messages);
+
+      // Verify original message wasn't mutated
+      expect(messages[0].content).toHaveLength(2);
+      expect(messages[0].content[1].type).toBe('image_url');
+      expect(messages[0].content[1].image_url.url).toBe('data:image/png;base64,ABC123');
+    });
+
+    it('should handle message window size correctly', async () => {
+      const { HumanMessage, AIMessage } = require('@langchain/core/messages');
+      const messages = [
+        new HumanMessage('Message 1'),
+        new AIMessage('Response 1'),
+        new HumanMessage('Message 2'),
+        new AIMessage('Response 2'),
+        new HumanMessage('Message 3'),
+        new AIMessage('Response 3'),
+      ];
+
+      // Window size is set to 3 in mockReq
+      await client.runMemory(messages);
+
+      expect(mockProcessMemory).toHaveBeenCalledTimes(1);
+      const processedMessage = mockProcessMemory.mock.calls[0][0][0];
+
+      // Should only include last 3 messages due to window size
+      expect(processedMessage.content).toContain('Message 3');
+      expect(processedMessage.content).toContain('Response 3');
+      expect(processedMessage.content).not.toContain('Message 1');
+      expect(processedMessage.content).not.toContain('Response 1');
+    });
+
+    it('should return early if processMemory is not set', async () => {
+      const { HumanMessage } = require('@langchain/core/messages');
+      client.processMemory = null;
+
+      const result = await client.runMemory([new HumanMessage('Test')]);
+
+      expect(result).toBeUndefined();
+      expect(mockProcessMemory).not.toHaveBeenCalled();
+    });
+  });
+});

api/server/controllers/agents/errors.js

@@ -1,10 +1,10 @@
 // errorHandler.js
-const { logger } = require('~/config');
-const getLogStores = require('~/cache/getLogStores');
+const { logger } = require('@librechat/data-schemas');
 const { CacheKeys, ViolationTypes } = require('librechat-data-provider');
+const { sendResponse } = require('~/server/middleware/error');
 const { recordUsage } = require('~/server/services/Threads');
 const { getConvo } = require('~/models/Conversation');
-const { sendResponse } = require('~/server/utils');
+const getLogStores = require('~/cache/getLogStores');
 
 /**
  * @typedef {Object} ErrorHandlerContext
@@ -75,7 +75,7 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
     } else if (/Files.*are invalid/.test(error.message)) {
       const errorMessage = `Files are invalid, or may not have uploaded yet.${
         endpoint === 'azureAssistants'
-          ? ' If using Azure OpenAI, files are only available in the region of the assistant\'s model at the time of upload.'
+          ? " If using Azure OpenAI, files are only available in the region of the assistant's model at the time of upload."
           : ''
       }`;
       return sendResponse(req, res, messageData, errorMessage);
@@ -105,8 +105,6 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
         return res.end();
       }
       await cache.delete(cacheKey);
-      // const cancelledRun = await openai.beta.threads.runs.cancel(thread_id, run_id);
-      // logger.debug(`[${originPath}] Cancelled run:`, cancelledRun);
     } catch (error) {
       logger.error(`[${originPath}] Error cancelling run`, error);
     }
@@ -115,7 +113,6 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
 
     let run;
     try {
-      // run = await openai.beta.threads.runs.retrieve(thread_id, run_id);
       await recordUsage({
         ...run.usage,
         model: run.model,
@@ -128,18 +125,9 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
 
     let finalEvent;
     try {
-      // const errorContentPart = {
-      //   text: {
-      //     value:
-      //       error?.message ?? 'There was an error processing your request. Please try again later.',
-      //   },
-      //   type: ContentTypes.ERROR,
-      // };
-
       finalEvent = {
         final: true,
         conversation: await getConvo(req.user.id, conversationId),
-        // runMessages,
       };
     } catch (error) {
       logger.error(`[${originPath}] Error finalizing error process`, error);

api/server/controllers/agents/llm.js

@@ -1,106 +0,0 @@
-const { HttpsProxyAgent } = require('https-proxy-agent');
-const { resolveHeaders } = require('librechat-data-provider');
-const { createLLM } = require('~/app/clients/llm');
-
-/**
- * Initializes and returns a Language Learning Model (LLM) instance.
- *
- * @param {Object} options - Configuration options for the LLM.
- * @param {string} options.model - The model identifier.
- * @param {string} options.modelName - The specific name of the model.
- * @param {number} options.temperature - The temperature setting for the model.
- * @param {number} options.presence_penalty - The presence penalty for the model.
- * @param {number} options.frequency_penalty - The frequency penalty for the model.
- * @param {number} options.max_tokens - The maximum number of tokens for the model output.
- * @param {boolean} options.streaming - Whether to use streaming for the model output.
- * @param {Object} options.context - The context for the conversation.
- * @param {number} options.tokenBuffer - The token buffer size.
- * @param {number} options.initialMessageCount - The initial message count.
- * @param {string} options.conversationId - The ID of the conversation.
- * @param {string} options.user - The user identifier.
- * @param {string} options.langchainProxy - The langchain proxy URL.
- * @param {boolean} options.useOpenRouter - Whether to use OpenRouter.
- * @param {Object} options.options - Additional options.
- * @param {Object} options.options.headers - Custom headers for the request.
- * @param {string} options.options.proxy - Proxy URL.
- * @param {Object} options.options.req - The request object.
- * @param {Object} options.options.res - The response object.
- * @param {boolean} options.options.debug - Whether to enable debug mode.
- * @param {string} options.apiKey - The API key for authentication.
- * @param {Object} options.azure - Azure-specific configuration.
- * @param {Object} options.abortController - The AbortController instance.
- * @returns {Object} The initialized LLM instance.
- */
-function initializeLLM(options) {
-  const {
-    model,
-    modelName,
-    temperature,
-    presence_penalty,
-    frequency_penalty,
-    max_tokens,
-    streaming,
-    user,
-    langchainProxy,
-    useOpenRouter,
-    options: { headers, proxy },
-    apiKey,
-    azure,
-  } = options;
-
-  const modelOptions = {
-    modelName: modelName || model,
-    temperature,
-    presence_penalty,
-    frequency_penalty,
-    user,
-  };
-
-  if (max_tokens) {
-    modelOptions.max_tokens = max_tokens;
-  }
-
-  const configOptions = {};
-
-  if (langchainProxy) {
-    configOptions.basePath = langchainProxy;
-  }
-
-  if (useOpenRouter) {
-    configOptions.basePath = 'https://openrouter.ai/api/v1';
-    configOptions.baseOptions = {
-      headers: {
-        'HTTP-Referer': 'https://librechat.ai',
-        'X-Title': 'LibreChat',
-      },
-    };
-  }
-
-  if (headers && typeof headers === 'object' && !Array.isArray(headers)) {
-    configOptions.baseOptions = {
-      headers: resolveHeaders({
-        ...headers,
-        ...configOptions?.baseOptions?.headers,
-      }),
-    };
-  }
-
-  if (proxy) {
-    configOptions.httpAgent = new HttpsProxyAgent(proxy);
-    configOptions.httpsAgent = new HttpsProxyAgent(proxy);
-  }
-
-  const llm = createLLM({
-    modelOptions,
-    configOptions,
-    openAIApiKey: apiKey,
-    azure,
-    streaming,
-  });
-
-  return llm;
-}
-
-module.exports = {
-  initializeLLM,
-};

api/server/controllers/agents/request.js

@@ -1,3 +1,5 @@
+const { sendEvent } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
 const { Constants } = require('librechat-data-provider');
 const {
   handleAbortError,
@@ -5,17 +7,19 @@ const {
   cleanupAbortController,
 } = require('~/server/middleware');
 const { disposeClient, clientRegistry, requestDataMap } = require('~/server/cleanup');
-const { sendMessage } = require('~/server/utils');
 const { saveMessage } = require('~/models');
-const { logger } = require('~/config');
 
 const AgentController = async (req, res, next, initializeClient, addTitle) => {
   let {
     text,
+    isRegenerate,
     endpointOption,
     conversationId,
+    isContinued = false,
+    editedContent = null,
     parentMessageId = null,
     overrideParentMessageId = null,
+    responseMessageId: editedResponseMessageId = null,
   } = req.body;
 
   let sender;
@@ -67,7 +71,7 @@ const AgentController = async (req, res, next, initializeClient, addTitle) => {
             handler();
           }
         } catch (e) {
-          // Ignore cleanup errors
+          logger.error('[AgentController] Error in cleanup handler', e);
         }
       }
     }
@@ -155,7 +159,7 @@ const AgentController = async (req, res, next, initializeClient, addTitle) => {
       try {
         res.removeListener('close', closeHandler);
       } catch (e) {
-        // Ignore
+        logger.error('[AgentController] Error removing close listener', e);
       }
     });
 
@@ -163,10 +167,15 @@ const AgentController = async (req, res, next, initializeClient, addTitle) => {
       user: userId,
       onStart,
       getReqData,
+      isContinued,
+      isRegenerate,
+      editedContent,
       conversationId,
       parentMessageId,
       abortController,
       overrideParentMessageId,
+      isEdited: !!editedContent,
+      responseMessageId: editedResponseMessageId,
       progressOptions: {
         res,
       },
@@ -206,7 +215,7 @@ const AgentController = async (req, res, next, initializeClient, addTitle) => {
       // Create a new response object with minimal copies
       const finalResponse = { ...response };
 
-      sendMessage(res, {
+      sendEvent(res, {
         final: true,
         conversation,
         title: conversation.title,

api/server/controllers/agents/v1.js

@@ -1,38 +1,49 @@
+const { z } = require('zod');
 const fs = require('fs').promises;
 const { nanoid } = require('nanoid');
-const { logger, PermissionBits } = require('@librechat/data-schemas');
+const { logger } = require('@librechat/data-schemas');
+const { agentCreateSchema, agentUpdateSchema } = require('@librechat/api');
 const {
   Tools,
   SystemRoles,
   FileSources,
+  ResourceType,
+  AccessRoleIds,
+  PrincipalType,
   EToolResources,
+  PermissionBits,
   actionDelimiter,
+  removeNullishValues,
 } = require('librechat-data-provider');
 const {
-  getAgent,
+  getListAgentsByAccess,
+  countPromotedAgents,
+  revertAgentVersion,
   createAgent,
   updateAgent,
   deleteAgent,
-  getListAgentsByAccess,
+  getAgent,
 } = require('~/models/Agent');
 const {
-  grantPermission,
-  findAccessibleResources,
   findPubliclyAccessibleResources,
+  findAccessibleResources,
   hasPublicPermission,
+  grantPermission,
 } = require('~/server/services/PermissionService');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
 const { resizeAvatar } = require('~/server/services/Files/images/avatar');
+const { getFileStrategy } = require('~/server/utils/getFileStrategy');
 const { refreshS3Url } = require('~/server/services/Files/S3/crud');
 const { filterFile } = require('~/server/services/Files/process');
 const { updateAction, getActions } = require('~/models/Action');
 const { getCachedTools } = require('~/server/services/Config');
-const { revertAgentVersion } = require('~/models/Agent');
 const { deleteFileByFilter } = require('~/models/File');
+const { getCategoriesWithCounts } = require('~/models');
 
 const systemTools = {
   [Tools.execute_code]: true,
   [Tools.file_search]: true,
+  [Tools.web_search]: true,
 };
 
 /**
@@ -41,13 +52,17 @@ const systemTools = {
  * @param {ServerRequest} req - The request object.
  * @param {AgentCreateParams} req.body - The request body.
  * @param {ServerResponse} res - The response object.
- * @returns {Agent} 201 - success response - application/json
+ * @returns {Promise<Agent>} 201 - success response - application/json
  */
 const createAgentHandler = async (req, res) => {
   try {
-    const { tools = [], provider, name, description, instructions, model, ...agentData } = req.body;
+    const validatedData = agentCreateSchema.parse(req.body);
+    const { tools = [], ...agentData } = removeNullishValues(validatedData);
+
     const { id: userId } = req.user;
 
+    agentData.id = `agent_${nanoid()}`;
+    agentData.author = userId;
     agentData.tools = [];
 
     const availableTools = await getCachedTools({ includeGlobal: true });
@@ -61,26 +76,16 @@ const createAgentHandler = async (req, res) => {
       }
     }
 
-    Object.assign(agentData, {
-      author: userId,
-      name,
-      description,
-      instructions,
-      provider,
-      model,
-    });
-
-    agentData.id = `agent_${nanoid()}`;
     const agent = await createAgent(agentData);
 
     // Automatically grant owner permissions to the creator
     try {
       await grantPermission({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: userId,
-        resourceType: 'agent',
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
-        accessRoleId: 'agent_owner',
+        accessRoleId: AccessRoleIds.AGENT_OWNER,
         grantedBy: userId,
       });
       logger.debug(
@@ -95,6 +100,10 @@ const createAgentHandler = async (req, res) => {
 
     res.status(201).json(agent);
   } catch (error) {
+    if (error instanceof z.ZodError) {
+      logger.error('[/Agents] Validation error', error.errors);
+      return res.status(400).json({ error: 'Invalid request data', details: error.errors });
+    }
     logger.error('[/Agents] Error creating agent', error);
     res.status(500).json({ error: error.message });
   }
@@ -141,7 +150,7 @@ const getAgentHandler = async (req, res, expandProperties = false) => {
 
     // Check if agent is public
     const isPublic = await hasPublicPermission({
-      resourceType: 'agent',
+      resourceType: ResourceType.AGENT,
       resourceId: agent._id,
       requiredPermissions: PermissionBits.VIEW,
     });
@@ -193,7 +202,8 @@ const getAgentHandler = async (req, res, expandProperties = false) => {
 const updateAgentHandler = async (req, res) => {
   try {
     const id = req.params.id;
-    const { _id, ...updateData } = req.body;
+    const validatedData = agentUpdateSchema.parse(req.body);
+    const { _id, ...updateData } = removeNullishValues(validatedData);
     const existingAgent = await getAgent({ id });
 
     if (!existingAgent) {
@@ -217,6 +227,11 @@ const updateAgentHandler = async (req, res) => {
 
     return res.json(updatedAgent);
   } catch (error) {
+    if (error instanceof z.ZodError) {
+      logger.error('[/Agents/:id] Validation error', error.errors);
+      return res.status(400).json({ error: 'Invalid request data', details: error.errors });
+    }
+
     logger.error('[/Agents/:id] Error updating Agent', error);
 
     if (error.statusCode === 409) {
@@ -259,6 +274,8 @@ const duplicateAgentHandler = async (req, res) => {
       createdAt: _createdAt,
       updatedAt: _updatedAt,
       tool_resources: _tool_resources = {},
+      versions: _versions,
+      __v: _v,
       ...cloneData
     } = agent;
     cloneData.name = `${agent.name} (${new Date().toLocaleString('en-US', {
@@ -327,11 +344,11 @@ const duplicateAgentHandler = async (req, res) => {
     // Automatically grant owner permissions to the duplicator
     try {
       await grantPermission({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: userId,
-        resourceType: 'agent',
+        resourceType: ResourceType.AGENT,
         resourceId: newAgent._id,
-        accessRoleId: 'agent_owner',
+        accessRoleId: AccessRoleIds.AGENT_OWNER,
         grantedBy: userId,
       });
       logger.debug(
@@ -389,21 +406,55 @@ const deleteAgentHandler = async (req, res) => {
 const getListAgentsHandler = async (req, res) => {
   try {
     const userId = req.user.id;
+    const { category, search, limit, cursor, promoted } = req.query;
+    let requiredPermission = req.query.requiredPermission;
+    if (typeof requiredPermission === 'string') {
+      requiredPermission = parseInt(requiredPermission, 10);
+      if (isNaN(requiredPermission)) {
+        requiredPermission = PermissionBits.VIEW;
+      }
+    } else if (typeof requiredPermission !== 'number') {
+      requiredPermission = PermissionBits.VIEW;
+    }
+    // Base filter
+    const filter = {};
 
+    // Handle category filter - only apply if category is defined
+    if (category !== undefined && category.trim() !== '') {
+      filter.category = category;
+    }
+
+    // Handle promoted filter - only from query param
+    if (promoted === '1') {
+      filter.is_promoted = true;
+    } else if (promoted === '0') {
+      filter.is_promoted = { $ne: true };
+    }
+
+    // Handle search filter
+    if (search && search.trim() !== '') {
+      filter.$or = [
+        { name: { $regex: search.trim(), $options: 'i' } },
+        { description: { $regex: search.trim(), $options: 'i' } },
+      ];
+    }
     // Get agent IDs the user has VIEW access to via ACL
     const accessibleIds = await findAccessibleResources({
       userId,
-      resourceType: 'agent',
-      requiredPermissions: PermissionBits.VIEW,
+      role: req.user.role,
+      resourceType: ResourceType.AGENT,
+      requiredPermissions: requiredPermission,
     });
     const publiclyAccessibleIds = await findPubliclyAccessibleResources({
-      resourceType: 'agent',
+      resourceType: ResourceType.AGENT,
       requiredPermissions: PermissionBits.VIEW,
     });
     // Use the new ACL-aware function
     const data = await getListAgentsByAccess({
       accessibleIds,
-      otherParams: {}, // Can add query params here if needed
+      otherParams: filter,
+      limit,
+      after: cursor,
     });
     if (data?.data?.length) {
       data.data = data.data.map((agent) => {
@@ -439,9 +490,25 @@ const uploadAgentAvatarHandler = async (req, res) => {
       return res.status(400).json({ message: 'Agent ID is required' });
     }
 
+    const isAdmin = req.user.role === SystemRoles.ADMIN;
+    const existingAgent = await getAgent({ id: agent_id });
+
+    if (!existingAgent) {
+      return res.status(404).json({ error: 'Agent not found' });
+    }
+
+    const isAuthor = existingAgent.author.toString() === req.user.id.toString();
+    const hasEditPermission = existingAgent.isCollaborative || isAdmin || isAuthor;
+
+    if (!hasEditPermission) {
+      return res.status(403).json({
+        error: 'You do not have permission to modify this non-collaborative agent',
+      });
+    }
+
     const buffer = await fs.readFile(req.file.path);
 
-    const fileStrategy = req.app.locals.fileStrategy;
+    const fileStrategy = getFileStrategy(req.app.locals, { isAvatar: true });
 
     const resizedBuffer = await resizeAvatar({
       userId: req.user.id,
@@ -461,14 +528,7 @@ const uploadAgentAvatarHandler = async (req, res) => {
       source: fileStrategy,
     };
 
-    let _avatar;
-    try {
-      const agent = await getAgent({ id: agent_id });
-      _avatar = agent.avatar;
-    } catch (error) {
-      logger.error('[/:agent_id/avatar] Error fetching agent', error);
-      _avatar = {};
-    }
+    let _avatar = existingAgent.avatar;
 
     if (_avatar && _avatar.source) {
       const { deleteFile } = getStrategyFunctions(_avatar.source);
@@ -545,7 +605,7 @@ const revertAgentVersionHandler = async (req, res) => {
       return res.status(404).json({ error: 'Agent not found' });
     }
 
-    const isAuthor = existingAgent.author.toString() === req.user.id;
+    const isAuthor = existingAgent.author.toString() === req.user.id.toString();
     const hasEditPermission = existingAgent.isCollaborative || isAdmin || isAuthor;
 
     if (!hasEditPermission) {
@@ -570,7 +630,48 @@ const revertAgentVersionHandler = async (req, res) => {
     res.status(500).json({ error: error.message });
   }
 };
+/**
+ * Get all agent categories with counts
+ *
+ * @param {Object} _req - Express request object (unused)
+ * @param {Object} res - Express response object
+ */
+const getAgentCategories = async (_req, res) => {
+  try {
+    const categories = await getCategoriesWithCounts();
+    const promotedCount = await countPromotedAgents();
+    const formattedCategories = categories.map((category) => ({
+      value: category.value,
+      label: category.label,
+      count: category.agentCount,
+      description: category.description,
+    }));
 
+    if (promotedCount > 0) {
+      formattedCategories.unshift({
+        value: 'promoted',
+        label: 'Promoted',
+        count: promotedCount,
+        description: 'Our recommended agents',
+      });
+    }
+
+    formattedCategories.push({
+      value: 'all',
+      label: 'All',
+      description: 'All available agents',
+    });
+
+    res.status(200).json(formattedCategories);
+  } catch (error) {
+    logger.error('[/Agents/Marketplace] Error fetching agent categories:', error);
+    res.status(500).json({
+      error: 'Failed to fetch agent categories',
+      userMessage: 'Unable to load categories. Please refresh the page.',
+      suggestion: 'Try refreshing the page or check your network connection',
+    });
+  }
+};
 module.exports = {
   createAgent: createAgentHandler,
   getAgent: getAgentHandler,
@@ -580,4 +681,5 @@ module.exports = {
   getListAgents: getListAgentsHandler,
   uploadAgentAvatar: uploadAgentAvatarHandler,
   revertAgentVersion: revertAgentVersionHandler,
+  getAgentCategories,
 };

api/server/controllers/assistants/chatV1.js

@@ -1,4 +1,7 @@
 const { v4 } = require('uuid');
+const { sleep } = require('@librechat/agents');
+const { sendEvent } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
 const {
   Time,
   Constants,
@@ -19,20 +22,20 @@ const {
   addThreadMetadata,
   saveAssistantMessage,
 } = require('~/server/services/Threads');
-const { sendResponse, sendMessage, sleep, countTokens } = require('~/server/utils');
 const { runAssistant, createOnTextProgress } = require('~/server/services/AssistantService');
 const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
 const { formatMessage, createVisionPrompt } = require('~/app/clients/prompts');
 const { createRun, StreamRunManager } = require('~/server/services/Runs');
 const { addTitle } = require('~/server/services/Endpoints/assistants');
 const { createRunBody } = require('~/server/services/createRunBody');
+const { sendResponse } = require('~/server/middleware/error');
 const { getTransactions } = require('~/models/Transaction');
 const { checkBalance } = require('~/models/balanceMethods');
 const { getConvo } = require('~/models/Conversation');
 const getLogStores = require('~/cache/getLogStores');
+const { countTokens } = require('~/server/utils');
 const { getModelMaxTokens } = require('~/utils');
 const { getOpenAIClient } = require('./helpers');
-const { logger } = require('~/config');
 
 /**
  * @route POST /
@@ -149,7 +152,7 @@ const chatV1 = async (req, res) => {
         return res.end();
       }
       await cache.delete(cacheKey);
-      const cancelledRun = await openai.beta.threads.runs.cancel(thread_id, run_id);
+      const cancelledRun = await openai.beta.threads.runs.cancel(run_id, { thread_id });
       logger.debug('[/assistants/chat/] Cancelled run:', cancelledRun);
     } catch (error) {
       logger.error('[/assistants/chat/] Error cancelling run', error);
@@ -159,7 +162,7 @@ const chatV1 = async (req, res) => {
 
     let run;
     try {
-      run = await openai.beta.threads.runs.retrieve(thread_id, run_id);
+      run = await openai.beta.threads.runs.retrieve(run_id, { thread_id });
       await recordUsage({
         ...run.usage,
         model: run.model,
@@ -471,7 +474,7 @@ const chatV1 = async (req, res) => {
     await Promise.all(promises);
 
     const sendInitialResponse = () => {
-      sendMessage(res, {
+      sendEvent(res, {
         sync: true,
         conversationId,
         // messages: previousMessages,
@@ -587,7 +590,7 @@ const chatV1 = async (req, res) => {
       iconURL: endpointOption.iconURL,
     };
 
-    sendMessage(res, {
+    sendEvent(res, {
       final: true,
       conversation,
       requestMessage: {
@@ -620,7 +623,7 @@ const chatV1 = async (req, res) => {
 
     if (!response.run.usage) {
       await sleep(3000);
-      completedRun = await openai.beta.threads.runs.retrieve(thread_id, response.run.id);
+      completedRun = await openai.beta.threads.runs.retrieve(response.run.id, { thread_id });
       if (completedRun.usage) {
         await recordUsage({
           ...completedRun.usage,

api/server/controllers/assistants/chatV2.js

@@ -1,4 +1,7 @@
 const { v4 } = require('uuid');
+const { sleep } = require('@librechat/agents');
+const { sendEvent } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
 const {
   Time,
   Constants,
@@ -22,15 +25,14 @@ const { createErrorHandler } = require('~/server/controllers/assistants/errors')
 const validateAuthor = require('~/server/middleware/assistants/validateAuthor');
 const { createRun, StreamRunManager } = require('~/server/services/Runs');
 const { addTitle } = require('~/server/services/Endpoints/assistants');
-const { sendMessage, sleep, countTokens } = require('~/server/utils');
 const { createRunBody } = require('~/server/services/createRunBody');
 const { getTransactions } = require('~/models/Transaction');
 const { checkBalance } = require('~/models/balanceMethods');
 const { getConvo } = require('~/models/Conversation');
 const getLogStores = require('~/cache/getLogStores');
+const { countTokens } = require('~/server/utils');
 const { getModelMaxTokens } = require('~/utils');
 const { getOpenAIClient } = require('./helpers');
-const { logger } = require('~/config');
 
 /**
  * @route POST /
@@ -309,7 +311,7 @@ const chatV2 = async (req, res) => {
     await Promise.all(promises);
 
     const sendInitialResponse = () => {
-      sendMessage(res, {
+      sendEvent(res, {
         sync: true,
         conversationId,
         // messages: previousMessages,
@@ -432,7 +434,7 @@ const chatV2 = async (req, res) => {
       iconURL: endpointOption.iconURL,
     };
 
-    sendMessage(res, {
+    sendEvent(res, {
       final: true,
       conversation,
       requestMessage: {
@@ -465,7 +467,7 @@ const chatV2 = async (req, res) => {
 
     if (!response.run.usage) {
       await sleep(3000);
-      completedRun = await openai.beta.threads.runs.retrieve(thread_id, response.run.id);
+      completedRun = await openai.beta.threads.runs.retrieve(response.run.id, { thread_id });
       if (completedRun.usage) {
         await recordUsage({
           ...completedRun.usage,

api/server/controllers/assistants/errors.js

@@ -1,10 +1,10 @@
 // errorHandler.js
-const { sendResponse } = require('~/server/utils');
-const { logger } = require('~/config');
-const getLogStores = require('~/cache/getLogStores');
+const { logger } = require('@librechat/data-schemas');
 const { CacheKeys, ViolationTypes, ContentTypes } = require('librechat-data-provider');
-const { getConvo } = require('~/models/Conversation');
 const { recordUsage, checkMessageGaps } = require('~/server/services/Threads');
+const { sendResponse } = require('~/server/middleware/error');
+const { getConvo } = require('~/models/Conversation');
+const getLogStores = require('~/cache/getLogStores');
 
 /**
  * @typedef {Object} ErrorHandlerContext
@@ -78,7 +78,7 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
     } else if (/Files.*are invalid/.test(error.message)) {
       const errorMessage = `Files are invalid, or may not have uploaded yet.${
         endpoint === 'azureAssistants'
-          ? ' If using Azure OpenAI, files are only available in the region of the assistant\'s model at the time of upload.'
+          ? " If using Azure OpenAI, files are only available in the region of the assistant's model at the time of upload."
           : ''
       }`;
       return sendResponse(req, res, messageData, errorMessage);
@@ -108,7 +108,7 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
         return res.end();
       }
       await cache.delete(cacheKey);
-      const cancelledRun = await openai.beta.threads.runs.cancel(thread_id, run_id);
+      const cancelledRun = await openai.beta.threads.runs.cancel(run_id, { thread_id });
       logger.debug(`[${originPath}] Cancelled run:`, cancelledRun);
     } catch (error) {
       logger.error(`[${originPath}] Error cancelling run`, error);
@@ -118,7 +118,7 @@ const createErrorHandler = ({ req, res, getContext, originPath = '/assistants/ch
 
     let run;
     try {
-      run = await openai.beta.threads.runs.retrieve(thread_id, run_id);
+      run = await openai.beta.threads.runs.retrieve(run_id, { thread_id });
       await recordUsage({
         ...run.usage,
         model: run.model,

api/server/controllers/assistants/helpers.js

@@ -173,6 +173,16 @@ const listAssistantsForAzure = async ({ req, res, version, azureConfig = {}, que
   };
 };
 
+/**
+ * Initializes the OpenAI client.
+ * @param {object} params - The parameters object.
+ * @param {ServerRequest} params.req - The request object.
+ * @param {ServerResponse} params.res - The response object.
+ * @param {TEndpointOption} params.endpointOption - The endpoint options.
+ * @param {boolean} params.initAppClient - Whether to initialize the app client.
+ * @param {string} params.overrideEndpoint - The endpoint to override.
+ * @returns {Promise<{ openai: OpenAIClient, openAIApiKey: string; client: import('~/app/clients/OpenAIClient') }>} - The initialized OpenAI client.
+ */
 async function getOpenAIClient({ req, res, endpointOption, initAppClient, overrideEndpoint }) {
   let endpoint = overrideEndpoint ?? req.body.endpoint ?? req.query.endpoint;
   const version = await getCurrentVersion(req, endpoint);

api/server/controllers/assistants/v1.js

@@ -197,7 +197,7 @@ const deleteAssistant = async (req, res) => {
     await validateAuthor({ req, openai });
 
     const assistant_id = req.params.id;
-    const deletionStatus = await openai.beta.assistants.del(assistant_id);
+    const deletionStatus = await openai.beta.assistants.delete(assistant_id);
     if (deletionStatus?.deleted) {
       await deleteAssistantActions({ req, assistant_id });
     }
@@ -365,7 +365,7 @@ const uploadAssistantAvatar = async (req, res) => {
     try {
       await fs.unlink(req.file.path);
       logger.debug('[/:agent_id/avatar] Temp. image upload file deleted');
-    } catch (error) {
+    } catch {
       logger.debug('[/:agent_id/avatar] Temp. image upload file already deleted');
     }
   }

api/server/controllers/tools.js

@@ -1,21 +1,21 @@
 const { nanoid } = require('nanoid');
 const { EnvVar } = require('@librechat/agents');
+const { logger } = require('@librechat/data-schemas');
+const { checkAccess, loadWebSearchAuth } = require('@librechat/api');
 const {
   Tools,
   AuthType,
   Permissions,
   ToolCallTypes,
   PermissionTypes,
-  loadWebSearchAuth,
 } = require('librechat-data-provider');
 const { processFileURL, uploadImageBuffer } = require('~/server/services/Files/process');
 const { processCodeOutput } = require('~/server/services/Files/Code/process');
 const { createToolCall, getToolCallsByConvo } = require('~/models/ToolCall');
 const { loadAuthValues } = require('~/server/services/Tools/credentials');
 const { loadTools } = require('~/app/clients/tools/util');
-const { checkAccess } = require('~/server/middleware');
+const { getRoleByName } = require('~/models/Role');
 const { getMessage } = require('~/models/Message');
-const { logger } = require('~/config');
 
 const fieldsMap = {
   [Tools.execute_code]: [EnvVar.CODE_API_KEY],
@@ -79,6 +79,7 @@ const verifyToolAuth = async (req, res) => {
         throwError: false,
       });
     } catch (error) {
+      logger.error('Error loading auth values', error);
       res.status(200).json({ authenticated: false, message: AuthType.USER_PROVIDED });
       return;
     }
@@ -132,7 +133,12 @@ const callTool = async (req, res) => {
     logger.debug(`[${toolId}/call] User: ${req.user.id}`);
     let hasAccess = true;
     if (toolAccessPermType[toolId]) {
-      hasAccess = await checkAccess(req.user, toolAccessPermType[toolId], [Permissions.USE]);
+      hasAccess = await checkAccess({
+        user: req.user,
+        permissionType: toolAccessPermType[toolId],
+        permissions: [Permissions.USE],
+        getRoleByName,
+      });
     }
     if (!hasAccess) {
       logger.warn(

api/server/index.js

@@ -16,7 +16,7 @@ const { connectDb, indexSync } = require('~/db');
 const validateImageRequest = require('./middleware/validateImageRequest');
 const { jwtLogin, ldapLogin, passportLogin } = require('~/strategies');
 const errorController = require('./controllers/ErrorController');
-const initializeMCP = require('./services/initializeMCP');
+const initializeMCPs = require('./services/initializeMCPs');
 const configureSocialLogins = require('./socialLogins');
 const AppService = require('./services/AppService');
 const staticCache = require('./utils/staticCache');
@@ -55,7 +55,6 @@ const startServer = async () => {
 
   /* Middleware */
   app.use(noIndex);
-  app.use(errorController);
   app.use(express.json({ limit: '3mb' }));
   app.use(express.urlencoded({ extended: true, limit: '3mb' }));
   app.use(mongoSanitize());
@@ -123,6 +122,9 @@ const startServer = async () => {
   app.use('/api/tags', routes.tags);
   app.use('/api/mcp', routes.mcp);
 
+  // Add the error controller one more time after all routes
+  app.use(errorController);
+
   app.use((req, res) => {
     res.set({
       'Cache-Control': process.env.INDEX_CACHE_CONTROL || 'no-cache, no-store, must-revalidate',
@@ -146,7 +148,7 @@ const startServer = async () => {
       logger.info(`Server listening at http://${host == '0.0.0.0' ? 'localhost' : host}:${port}`);
     }
 
-    initializeMCP(app);
+    initializeMCPs(app);
   });
 };
 

api/server/index.spec.js

@@ -1,5 +1,4 @@
 const fs = require('fs');
-const path = require('path');
 const request = require('supertest');
 const { MongoMemoryServer } = require('mongodb-memory-server');
 const mongoose = require('mongoose');
@@ -59,6 +58,30 @@ describe('Server Configuration', () => {
     expect(response.headers['pragma']).toBe('no-cache');
     expect(response.headers['expires']).toBe('0');
   });
+
+  it('should return 500 for unknown errors via ErrorController', async () => {
+    // Testing the error handling here on top of unit tests to ensure the middleware is correctly integrated
+
+    // Mock MongoDB operations to fail
+    const originalFindOne = mongoose.models.User.findOne;
+    const mockError = new Error('MongoDB operation failed');
+    mongoose.models.User.findOne = jest.fn().mockImplementation(() => {
+      throw mockError;
+    });
+
+    try {
+      const response = await request(app).post('/api/auth/login').send({
+        email: 'test@example.com',
+        password: 'password123',
+      });
+
+      expect(response.status).toBe(500);
+      expect(response.text).toBe('An unknown error occurred.');
+    } finally {
+      // Restore original function
+      mongoose.models.User.findOne = originalFindOne;
+    }
+  });
 });
 
 // Polls the /health endpoint every 30ms for up to 10 seconds to wait for the server to start completely

api/server/middleware/abortMiddleware.js

@@ -1,13 +1,13 @@
-// abortMiddleware.js
+const { logger } = require('@librechat/data-schemas');
+const { countTokens, isEnabled, sendEvent } = require('@librechat/api');
 const { isAssistantsEndpoint, ErrorTypes } = require('librechat-data-provider');
-const { sendMessage, sendError, countTokens, isEnabled } = require('~/server/utils');
 const { truncateText, smartTruncateText } = require('~/app/clients/prompts');
 const clearPendingReq = require('~/cache/clearPendingReq');
+const { sendError } = require('~/server/middleware/error');
 const { spendTokens } = require('~/models/spendTokens');
 const abortControllers = require('./abortControllers');
 const { saveMessage, getConvo } = require('~/models');
 const { abortRun } = require('./abortRun');
-const { logger } = require('~/config');
 
 const abortDataMap = new WeakMap();
 
@@ -101,7 +101,7 @@ async function abortMessage(req, res) {
   cleanupAbortController(abortKey);
 
   if (res.headersSent && finalEvent) {
-    return sendMessage(res, finalEvent);
+    return sendEvent(res, finalEvent);
   }
 
   res.setHeader('Content-Type', 'application/json');
@@ -174,7 +174,7 @@ const createAbortController = (req, res, getAbortData, getReqData) => {
    * @param {string} responseMessageId
    */
   const onStart = (userMessage, responseMessageId) => {
-    sendMessage(res, { message: userMessage, created: true });
+    sendEvent(res, { message: userMessage, created: true });
 
     const abortKey = userMessage?.conversationId ?? req.user.id;
     getReqData({ abortKey });

api/server/middleware/abortRun.js

@@ -1,11 +1,11 @@
+const { sendEvent } = require('@librechat/api');
+const { logger } = require('@librechat/data-schemas');
 const { CacheKeys, RunStatus, isUUID } = require('librechat-data-provider');
 const { initializeClient } = require('~/server/services/Endpoints/assistants');
 const { checkMessageGaps, recordUsage } = require('~/server/services/Threads');
 const { deleteMessages } = require('~/models/Message');
 const { getConvo } = require('~/models/Conversation');
 const getLogStores = require('~/cache/getLogStores');
-const { sendMessage } = require('~/server/utils');
-const { logger } = require('~/config');
 
 const three_minutes = 1000 * 60 * 3;
 
@@ -34,7 +34,7 @@ async function abortRun(req, res) {
   const [thread_id, run_id] = runValues.split(':');
 
   if (!run_id) {
-    logger.warn('[abortRun] Couldn\'t find run for cancel request', { thread_id });
+    logger.warn("[abortRun] Couldn't find run for cancel request", { thread_id });
     return res.status(204).send({ message: 'Run not found' });
   } else if (run_id === 'cancelled') {
     logger.warn('[abortRun] Run already cancelled', { thread_id });
@@ -47,7 +47,7 @@ async function abortRun(req, res) {
 
   try {
     await cache.set(cacheKey, 'cancelled', three_minutes);
-    const cancelledRun = await openai.beta.threads.runs.cancel(thread_id, run_id);
+    const cancelledRun = await openai.beta.threads.runs.cancel(run_id, { thread_id });
     logger.debug('[abortRun] Cancelled run:', cancelledRun);
   } catch (error) {
     logger.error('[abortRun] Error cancelling run', error);
@@ -60,7 +60,7 @@ async function abortRun(req, res) {
   }
 
   try {
-    const run = await openai.beta.threads.runs.retrieve(thread_id, run_id);
+    const run = await openai.beta.threads.runs.retrieve(run_id, { thread_id });
     await recordUsage({
       ...run.usage,
       model: run.model,
@@ -93,7 +93,7 @@ async function abortRun(req, res) {
   };
 
   if (res.headersSent && finalEvent) {
-    return sendMessage(res, finalEvent);
+    return sendEvent(res, finalEvent);
   }
 
   res.json(finalEvent);

api/server/middleware/accessResources/canAccessAgentFromBody.js

@@ -1,5 +1,5 @@
 const { logger } = require('@librechat/data-schemas');
-const { Constants, isAgentsEndpoint } = require('librechat-data-provider');
+const { Constants, isAgentsEndpoint, ResourceType } = require('librechat-data-provider');
 const { canAccessResource } = require('./canAccessResource');
 const { getAgent } = require('~/models/Agent');
 
@@ -67,7 +67,7 @@ const canAccessAgentFromBody = (options) => {
       }
 
       const agentAccessMiddleware = canAccessResource({
-        resourceType: 'agent',
+        resourceType: ResourceType.AGENT,
         requiredPermission,
         resourceIdParam: 'agent_id', // This will be ignored since we use custom resolver
         idResolver: () => resolveAgentIdFromBody(agentId),

api/server/middleware/accessResources/canAccessAgentResource.js

@@ -1,5 +1,6 @@
-const { getAgent } = require('~/models/Agent');
+const { ResourceType } = require('librechat-data-provider');
 const { canAccessResource } = require('./canAccessResource');
+const { getAgent } = require('~/models/Agent');
 
 /**
  * Agent ID resolver function
@@ -46,7 +47,7 @@ const canAccessAgentResource = (options) => {
   }
 
   return canAccessResource({
-    resourceType: 'agent',
+    resourceType: ResourceType.AGENT,
     requiredPermission,
     resourceIdParam,
     idResolver: resolveAgentId,

api/server/middleware/accessResources/canAccessAgentResource.spec.js

@@ -1,4 +1,5 @@
 const mongoose = require('mongoose');
+const { ResourceType, PrincipalType, PrincipalModel } = require('librechat-data-provider');
 const { MongoMemoryServer } = require('mongodb-memory-server');
 const { canAccessAgentResource } = require('./canAccessAgentResource');
 const { User, Role, AclEntry } = require('~/db/models');
@@ -42,7 +43,7 @@ describe('canAccessAgentResource middleware', () => {
     });
 
     req = {
-      user: { id: testUser._id.toString(), role: 'test-role' },
+      user: { id: testUser._id, role: testUser.role },
       params: {},
     };
     res = {
@@ -96,10 +97,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry for the author (owner permissions)
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 15, // All permissions (1+2+4+8)
         grantedBy: testUser._id,
@@ -133,10 +134,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry for the other user (owner)
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: otherUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 15, // All permissions
         grantedBy: otherUser._id,
@@ -174,10 +175,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry granting view permission to test user
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 1, // VIEW permission
         grantedBy: otherUser._id,
@@ -211,10 +212,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry granting only view permission
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 1, // VIEW permission only
         grantedBy: otherUser._id,
@@ -258,10 +259,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry for the author
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 15, // All permissions
         grantedBy: testUser._id,
@@ -294,10 +295,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry with all permissions for the owner
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 15, // All permissions (1+2+4+8)
         grantedBy: testUser._id,
@@ -354,10 +355,10 @@ describe('canAccessAgentResource middleware', () => {
 
       // Create ACL entry for the author
       await AclEntry.create({
-        principalType: 'user',
+        principalType: PrincipalType.USER,
         principalId: testUser._id,
-        principalModel: 'User',
-        resourceType: 'agent',
+        principalModel: PrincipalModel.USER,
+        resourceType: ResourceType.AGENT,
         resourceId: agent._id,
         permBits: 15, // All permissions
         grantedBy: testUser._id,

api/server/middleware/accessResources/canAccessPromptGroupResource.js

@@ -0,0 +1,61 @@
+const { ResourceType } = require('librechat-data-provider');
+const { canAccessResource } = require('./canAccessResource');
+const { getPromptGroup } = require('~/models/Prompt');
+
+/**
+ * PromptGroup ID resolver function
+ * Resolves promptGroup ID to MongoDB ObjectId
+ *
+ * @param {string} groupId - PromptGroup ID from route parameter
+ * @returns {Promise<Object|null>} PromptGroup document with _id field, or null if not found
+ */
+const resolvePromptGroupId = async (groupId) => {
+  return await getPromptGroup({ _id: groupId });
+};
+
+/**
+ * PromptGroup-specific middleware factory that creates middleware to check promptGroup access permissions.
+ * This middleware extends the generic canAccessResource to handle promptGroup ID resolution.
+ *
+ * @param {Object} options - Configuration options
+ * @param {number} options.requiredPermission - The permission bit required (1=view, 2=edit, 4=delete, 8=share)
+ * @param {string} [options.resourceIdParam='groupId'] - The name of the route parameter containing the promptGroup ID
+ * @returns {Function} Express middleware function
+ *
+ * @example
+ * // Basic usage for viewing promptGroups
+ * router.get('/prompts/groups/:groupId',
+ *   canAccessPromptGroupResource({ requiredPermission: 1 }),
+ *   getPromptGroup
+ * );
+ *
+ * @example
+ * // Custom resource ID parameter and edit permission
+ * router.patch('/prompts/groups/:id',
+ *   canAccessPromptGroupResource({
+ *     requiredPermission: 2,
+ *     resourceIdParam: 'id'
+ *   }),
+ *   updatePromptGroup
+ * );
+ */
+const canAccessPromptGroupResource = (options) => {
+  const { requiredPermission, resourceIdParam = 'groupId' } = options;
+
+  if (!requiredPermission || typeof requiredPermission !== 'number') {
+    throw new Error(
+      'canAccessPromptGroupResource: requiredPermission is required and must be a number',
+    );
+  }
+
+  return canAccessResource({
+    resourceType: ResourceType.PROMPTGROUP,
+    requiredPermission,
+    resourceIdParam,
+    idResolver: resolvePromptGroupId,
+  });
+};
+
+module.exports = {
+  canAccessPromptGroupResource,
+};

api/server/middleware/accessResources/canAccessPromptViaGroup.js

@@ -0,0 +1,55 @@
+const { ResourceType } = require('librechat-data-provider');
+const { canAccessResource } = require('./canAccessResource');
+const { getPrompt } = require('~/models/Prompt');
+
+/**
+ * Prompt to PromptGroup ID resolver function
+ * Resolves prompt ID to its parent promptGroup ID
+ *
+ * @param {string} promptId - Prompt ID from route parameter
+ * @returns {Promise<Object|null>} Object with promptGroup's _id field, or null if not found
+ */
+const resolvePromptToGroupId = async (promptId) => {
+  const prompt = await getPrompt({ _id: promptId });
+  if (!prompt || !prompt.groupId) {
+    return null;
+  }
+  // Return an object with _id that matches the promptGroup ID
+  return { _id: prompt.groupId };
+};
+
+/**
+ * Middleware factory that checks promptGroup permissions when accessing individual prompts.
+ * This allows permission management at the promptGroup level while still supporting
+ * individual prompt access patterns.
+ *
+ * @param {Object} options - Configuration options
+ * @param {number} options.requiredPermission - The permission bit required (1=view, 2=edit, 4=delete, 8=share)
+ * @param {string} [options.resourceIdParam='promptId'] - The name of the route parameter containing the prompt ID
+ * @returns {Function} Express middleware function
+ *
+ * @example
+ * // Check promptGroup permissions when viewing a prompt
+ * router.get('/prompts/:promptId',
+ *   canAccessPromptViaGroup({ requiredPermission: 1 }),
+ *   getPrompt
+ * );
+ */
+const canAccessPromptViaGroup = (options) => {
+  const { requiredPermission, resourceIdParam = 'promptId' } = options;
+
+  if (!requiredPermission || typeof requiredPermission !== 'number') {
+    throw new Error('canAccessPromptViaGroup: requiredPermission is required and must be a number');
+  }
+
+  return canAccessResource({
+    resourceType: ResourceType.PROMPTGROUP,
+    requiredPermission,
+    resourceIdParam,
+    idResolver: resolvePromptToGroupId,
+  });
+};
+
+module.exports = {
+  canAccessPromptViaGroup,
+};

api/server/middleware/accessResources/canAccessResource.js

@@ -111,6 +111,7 @@ const canAccessResource = (options) => {
       // Check permissions using PermissionService with ObjectId
       const hasPermission = await checkPermission({
         userId,
+        role: req.user.role,
         resourceType,
         resourceId,
         requiredPermission,

api/server/middleware/accessResources/fileAccess.js

@@ -0,0 +1,125 @@
+const { logger } = require('@librechat/data-schemas');
+const { PermissionBits, hasPermissions, ResourceType } = require('librechat-data-provider');
+const { getEffectivePermissions } = require('~/server/services/PermissionService');
+const { getAgent } = require('~/models/Agent');
+const { getFiles } = require('~/models/File');
+
+/**
+ * Checks if user has access to a file through agent permissions
+ * Files inherit permissions from agents - if you can view the agent, you can access its files
+ */
+const checkAgentBasedFileAccess = async ({ userId, role, fileId }) => {
+  try {
+    // Find agents that have this file in their tool_resources
+    const agentsWithFile = await getAgent({
+      $or: [
+        { 'tool_resources.file_search.file_ids': fileId },
+        { 'tool_resources.execute_code.file_ids': fileId },
+        { 'tool_resources.ocr.file_ids': fileId },
+      ],
+    });
+
+    if (!agentsWithFile || agentsWithFile.length === 0) {
+      return false;
+    }
+
+    // Check if user has access to any of these agents
+    for (const agent of Array.isArray(agentsWithFile) ? agentsWithFile : [agentsWithFile]) {
+      // Check if user is the agent author
+      if (agent.author && agent.author.toString() === userId) {
+        logger.debug(`[fileAccess] User is author of agent ${agent.id}`);
+        return true;
+      }
+
+      // Check ACL permissions for VIEW access on the agent
+      try {
+        const permissions = await getEffectivePermissions({
+          userId,
+          role,
+          resourceType: ResourceType.AGENT,
+          resourceId: agent._id || agent.id,
+        });
+
+        if (hasPermissions(permissions, PermissionBits.VIEW)) {
+          logger.debug(`[fileAccess] User ${userId} has VIEW permissions on agent ${agent.id}`);
+          return true;
+        }
+      } catch (permissionError) {
+        logger.warn(
+          `[fileAccess] Permission check failed for agent ${agent.id}:`,
+          permissionError.message,
+        );
+        // Continue checking other agents
+      }
+    }
+
+    return false;
+  } catch (error) {
+    logger.error('[fileAccess] Error checking agent-based access:', error);
+    return false;
+  }
+};
+
+/**
+ * Middleware to check if user can access a file
+ * Checks: 1) File ownership, 2) Agent-based access (file inherits agent permissions)
+ */
+const fileAccess = async (req, res, next) => {
+  try {
+    const fileId = req.params.file_id;
+    const userId = req.user?.id;
+    const userRole = req.user?.role;
+    if (!fileId) {
+      return res.status(400).json({
+        error: 'Bad Request',
+        message: 'file_id is required',
+      });
+    }
+
+    if (!userId) {
+      return res.status(401).json({
+        error: 'Unauthorized',
+        message: 'Authentication required',
+      });
+    }
+
+    // Get the file
+    const [file] = await getFiles({ file_id: fileId });
+    if (!file) {
+      return res.status(404).json({
+        error: 'Not Found',
+        message: 'File not found',
+      });
+    }
+
+    // Check if user owns the file
+    if (file.user && file.user.toString() === userId) {
+      req.fileAccess = { file };
+      return next();
+    }
+
+    // Check agent-based access (file inherits agent permissions)
+    const hasAgentAccess = await checkAgentBasedFileAccess({ userId, role: userRole, fileId });
+    if (hasAgentAccess) {
+      req.fileAccess = { file };
+      return next();
+    }
+
+    // No access
+    logger.warn(`[fileAccess] User ${userId} denied access to file ${fileId}`);
+    return res.status(403).json({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to access this file',
+    });
+  } catch (error) {
+    logger.error('[fileAccess] Error checking file access:', error);
+    return res.status(500).json({
+      error: 'Internal Server Error',
+      message: 'Failed to check file access permissions',
+    });
+  }
+};
+
+module.exports = {
+  fileAccess,
+};

api/server/middleware/accessResources/index.js

@@ -1,9 +1,13 @@
 const { canAccessResource } = require('./canAccessResource');
 const { canAccessAgentResource } = require('./canAccessAgentResource');
 const { canAccessAgentFromBody } = require('./canAccessAgentFromBody');
+const { canAccessPromptViaGroup } = require('./canAccessPromptViaGroup');
+const { canAccessPromptGroupResource } = require('./canAccessPromptGroupResource');
 
 module.exports = {
   canAccessResource,
   canAccessAgentResource,
   canAccessAgentFromBody,
+  canAccessPromptViaGroup,
+  canAccessPromptGroupResource,
 };

api/server/middleware/buildEndpointOption.js

@@ -1,3 +1,4 @@
+const { handleError } = require('@librechat/api');
 const { logger } = require('@librechat/data-schemas');
 const {
   EndpointURLs,
@@ -7,7 +8,6 @@ const {
 } = require('librechat-data-provider');
 const azureAssistants = require('~/server/services/Endpoints/azureAssistants');
 const assistants = require('~/server/services/Endpoints/assistants');
-const gptPlugins = require('~/server/services/Endpoints/gptPlugins');
 const { processFiles } = require('~/server/services/Files/process');
 const anthropic = require('~/server/services/Endpoints/anthropic');
 const bedrock = require('~/server/services/Endpoints/bedrock');
@@ -15,7 +15,6 @@ const openAI = require('~/server/services/Endpoints/openAI');
 const agents = require('~/server/services/Endpoints/agents');
 const custom = require('~/server/services/Endpoints/custom');
 const google = require('~/server/services/Endpoints/google');
-const { handleError } = require('~/server/utils');
 
 const buildFunction = {
   [EModelEndpoint.openAI]: openAI.buildOptions,
@@ -25,7 +24,6 @@ const buildFunction = {
   [EModelEndpoint.bedrock]: bedrock.buildOptions,
   [EModelEndpoint.azureOpenAI]: openAI.buildOptions,
   [EModelEndpoint.anthropic]: anthropic.buildOptions,
-  [EModelEndpoint.gptPlugins]: gptPlugins.buildOptions,
   [EModelEndpoint.assistants]: assistants.buildOptions,
   [EModelEndpoint.azureAssistants]: azureAssistants.buildOptions,
 };
@@ -60,15 +58,6 @@ async function buildEndpointOption(req, res, next) {
       return handleError(res, { text: 'Model spec mismatch' });
     }
 
-    if (
-      currentModelSpec.preset.endpoint !== EModelEndpoint.gptPlugins &&
-      currentModelSpec.preset.tools
-    ) {
-      return handleError(res, {
-        text: `Only the "${EModelEndpoint.gptPlugins}" endpoint can have tools defined in the preset`,
-      });
-    }
-
     try {
       currentModelSpec.preset.spec = spec;
       if (currentModelSpec.iconURL != null && currentModelSpec.iconURL !== '') {

api/server/middleware/checkBan.js

@@ -18,7 +18,6 @@ const message = 'Your account has been temporarily banned due to violations of o
  * @function
  * @param {Object} req - Express Request object.
  * @param {Object} res - Express Response object.
- * @param {String} errorMessage - Error message to be displayed in case of /api/ask or /api/edit request.
  *
  * @returns {Promise<Object>} - Returns a Promise which when resolved sends a response status of 403 with a specific message if request is not of api/ask or api/edit types. If it is, calls `denyRequest()` function.
  */
@@ -135,6 +134,7 @@ const checkBan = async (req, res, next = () => {}) => {
     return await banResponse(req, res);
   } catch (error) {
     logger.error('Error in checkBan middleware:', error);
+    return next(error);
   }
 };
 

api/server/middleware/checkPeoplePickerAccess.js

@@ -0,0 +1,82 @@
+const { PrincipalType, PermissionTypes, Permissions } = require('librechat-data-provider');
+const { getRoleByName } = require('~/models/Role');
+const { logger } = require('~/config');
+
+/**
+ * Middleware to check if user has permission to access people picker functionality
+ * Checks specific permission based on the 'type' query parameter:
+ * - type=user: requires VIEW_USERS permission
+ * - type=group: requires VIEW_GROUPS permission
+ * - type=role: requires VIEW_ROLES permission
+ * - no type (mixed search): requires either VIEW_USERS OR VIEW_GROUPS OR VIEW_ROLES
+ */
+const checkPeoplePickerAccess = async (req, res, next) => {
+  try {
+    const user = req.user;
+    if (!user || !user.role) {
+      return res.status(401).json({
+        error: 'Unauthorized',
+        message: 'Authentication required',
+      });
+    }
+
+    const role = await getRoleByName(user.role);
+    if (!role || !role.permissions) {
+      return res.status(403).json({
+        error: 'Forbidden',
+        message: 'No permissions configured for user role',
+      });
+    }
+
+    const { type } = req.query;
+    const peoplePickerPerms = role.permissions[PermissionTypes.PEOPLE_PICKER] || {};
+    const canViewUsers = peoplePickerPerms[Permissions.VIEW_USERS] === true;
+    const canViewGroups = peoplePickerPerms[Permissions.VIEW_GROUPS] === true;
+    const canViewRoles = peoplePickerPerms[Permissions.VIEW_ROLES] === true;
+
+    const permissionChecks = {
+      [PrincipalType.USER]: {
+        hasPermission: canViewUsers,
+        message: 'Insufficient permissions to search for users',
+      },
+      [PrincipalType.GROUP]: {
+        hasPermission: canViewGroups,
+        message: 'Insufficient permissions to search for groups',
+      },
+      [PrincipalType.ROLE]: {
+        hasPermission: canViewRoles,
+        message: 'Insufficient permissions to search for roles',
+      },
+    };
+
+    const check = permissionChecks[type];
+    if (check && !check.hasPermission) {
+      return res.status(403).json({
+        error: 'Forbidden',
+        message: check.message,
+      });
+    }
+
+    if (!type && !canViewUsers && !canViewGroups && !canViewRoles) {
+      return res.status(403).json({
+        error: 'Forbidden',
+        message: 'Insufficient permissions to search for users, groups, or roles',
+      });
+    }
+
+    next();
+  } catch (error) {
+    logger.error(
+      `[checkPeoplePickerAccess][${req.user?.id}] checkPeoplePickerAccess error for req.query.type = ${req.query.type}`,
+      error,
+    );
+    return res.status(500).json({
+      error: 'Internal Server Error',
+      message: 'Failed to check permissions',
+    });
+  }
+};
+
+module.exports = {
+  checkPeoplePickerAccess,
+};

api/server/middleware/checkPeoplePickerAccess.spec.js

@@ -0,0 +1,250 @@
+const { PrincipalType, PermissionTypes, Permissions } = require('librechat-data-provider');
+const { checkPeoplePickerAccess } = require('./checkPeoplePickerAccess');
+const { getRoleByName } = require('~/models/Role');
+const { logger } = require('~/config');
+
+jest.mock('~/models/Role');
+jest.mock('~/config', () => ({
+  logger: {
+    error: jest.fn(),
+  },
+}));
+
+describe('checkPeoplePickerAccess', () => {
+  let req, res, next;
+
+  beforeEach(() => {
+    req = {
+      user: { id: 'user123', role: 'USER' },
+      query: {},
+    };
+    res = {
+      status: jest.fn().mockReturnThis(),
+      json: jest.fn(),
+    };
+    next = jest.fn();
+    jest.clearAllMocks();
+  });
+
+  it('should return 401 if user is not authenticated', async () => {
+    req.user = null;
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(401);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Unauthorized',
+      message: 'Authentication required',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should return 403 if role has no permissions', async () => {
+    getRoleByName.mockResolvedValue(null);
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'No permissions configured for user role',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should allow access when searching for users with VIEW_USERS permission', async () => {
+    req.query.type = PrincipalType.USER;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: true,
+          [Permissions.VIEW_GROUPS]: false,
+          [Permissions.VIEW_ROLES]: false,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(next).toHaveBeenCalled();
+    expect(res.status).not.toHaveBeenCalled();
+  });
+
+  it('should deny access when searching for users without VIEW_USERS permission', async () => {
+    req.query.type = PrincipalType.USER;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: false,
+          [Permissions.VIEW_GROUPS]: true,
+          [Permissions.VIEW_ROLES]: true,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to search for users',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should allow access when searching for groups with VIEW_GROUPS permission', async () => {
+    req.query.type = PrincipalType.GROUP;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: false,
+          [Permissions.VIEW_GROUPS]: true,
+          [Permissions.VIEW_ROLES]: false,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(next).toHaveBeenCalled();
+    expect(res.status).not.toHaveBeenCalled();
+  });
+
+  it('should deny access when searching for groups without VIEW_GROUPS permission', async () => {
+    req.query.type = PrincipalType.GROUP;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: true,
+          [Permissions.VIEW_GROUPS]: false,
+          [Permissions.VIEW_ROLES]: true,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to search for groups',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should allow access when searching for roles with VIEW_ROLES permission', async () => {
+    req.query.type = PrincipalType.ROLE;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: false,
+          [Permissions.VIEW_GROUPS]: false,
+          [Permissions.VIEW_ROLES]: true,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(next).toHaveBeenCalled();
+    expect(res.status).not.toHaveBeenCalled();
+  });
+
+  it('should deny access when searching for roles without VIEW_ROLES permission', async () => {
+    req.query.type = PrincipalType.ROLE;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: true,
+          [Permissions.VIEW_GROUPS]: true,
+          [Permissions.VIEW_ROLES]: false,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to search for roles',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should allow mixed search when user has at least one permission', async () => {
+    // No type specified = mixed search
+    req.query.type = undefined;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: false,
+          [Permissions.VIEW_GROUPS]: false,
+          [Permissions.VIEW_ROLES]: true,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(next).toHaveBeenCalled();
+    expect(res.status).not.toHaveBeenCalled();
+  });
+
+  it('should deny mixed search when user has no permissions', async () => {
+    // No type specified = mixed search
+    req.query.type = undefined;
+    getRoleByName.mockResolvedValue({
+      permissions: {
+        [PermissionTypes.PEOPLE_PICKER]: {
+          [Permissions.VIEW_USERS]: false,
+          [Permissions.VIEW_GROUPS]: false,
+          [Permissions.VIEW_ROLES]: false,
+        },
+      },
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to search for users, groups, or roles',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should handle errors gracefully', async () => {
+    const error = new Error('Database error');
+    getRoleByName.mockRejectedValue(error);
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(logger.error).toHaveBeenCalledWith(
+      '[checkPeoplePickerAccess][user123] checkPeoplePickerAccess error for req.query.type = undefined',
+      error,
+    );
+    expect(res.status).toHaveBeenCalledWith(500);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Internal Server Error',
+      message: 'Failed to check permissions',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+
+  it('should handle missing permissions object gracefully', async () => {
+    req.query.type = PrincipalType.USER;
+    getRoleByName.mockResolvedValue({
+      permissions: {}, // No PEOPLE_PICKER permissions
+    });
+
+    await checkPeoplePickerAccess(req, res, next);
+
+    expect(res.status).toHaveBeenCalledWith(403);
+    expect(res.json).toHaveBeenCalledWith({
+      error: 'Forbidden',
+      message: 'Insufficient permissions to search for users',
+    });
+    expect(next).not.toHaveBeenCalled();
+  });
+});

api/server/middleware/concurrentLimiter.js

@@ -1,4 +1,4 @@
-const { Time, CacheKeys } = require('librechat-data-provider');
+const { Time, CacheKeys, ViolationTypes } = require('librechat-data-provider');
 const clearPendingReq = require('~/cache/clearPendingReq');
 const { logViolation, getLogStores } = require('~/cache');
 const { isEnabled } = require('~/server/utils');
@@ -37,7 +37,7 @@ const concurrentLimiter = async (req, res, next) => {
 
   const userId = req.user?.id ?? req.user?._id ?? '';
   const limit = Math.max(CONCURRENT_MESSAGE_MAX, 1);
-  const type = 'concurrent';
+  const type = ViolationTypes.CONCURRENT;
 
   const key = `${isEnabled(USE_REDIS) ? namespace : ''}:${userId}`;
   const pendingRequests = +((await cache.get(key)) ?? 0);

api/server/middleware/denyRequest.js

@@ -1,6 +1,7 @@
 const crypto = require('crypto');
+const { sendEvent } = require('@librechat/api');
 const { getResponseSender, Constants } = require('librechat-data-provider');
-const { sendMessage, sendError } = require('~/server/utils');
+const { sendError } = require('~/server/middleware/error');
 const { saveMessage } = require('~/models');
 
 /**
@@ -36,7 +37,7 @@ const denyRequest = async (req, res, errorMessage) => {
     isCreatedByUser: true,
     text,
   };
-  sendMessage(res, { message: userMessage, created: true });
+  sendEvent(res, { message: userMessage, created: true });
 
   const shouldSaveMessage = _convoId && parentMessageId && parentMessageId !== Constants.NO_PARENT;