refactor: correct type definitions for ToolInputSchema and error handling in searchFiles function

cleanup: code formatting and improve readability across multiple components
2025-05-31 20:53:12 +02:00 · 2025-05-31 20:48:23 +02:00
1800 changed files with 33811 additions and 149880 deletions
--- a/.env.example
+++ b/.env.example
@@ -15,20 +15,6 @@ 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
@@ -40,13 +26,6 @@ NO_INDEX=true
 # Defaulted to 1.
 TRUST_PROXY=1

-# Minimum password length for user authentication
-# Default: 8
-# Note: When using LDAP authentication, you may want to set this to 1 
-# to bypass local password validation, as LDAP servers handle their own
-# password policies.
-# MIN_PASSWORD_LENGTH=8
-
 #===============#
 # JSON Logging  #
 #===============#
@@ -79,7 +58,7 @@ DEBUG_CONSOLE=false
 #                     Endpoints                     #
 #===================================================#

-# ENDPOINTS=openAI,assistants,azureOpenAI,google,anthropic
+# ENDPOINTS=openAI,assistants,azureOpenAI,google,gptPlugins,anthropic

 PROXY=

@@ -163,10 +142,10 @@ GOOGLE_KEY=user_provided
 # GOOGLE_AUTH_HEADER=true

 # Gemini API (AI Studio)
-# GOOGLE_MODELS=gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite,gemini-2.0-flash,gemini-2.0-flash-lite
+# 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

 # Vertex AI
-# GOOGLE_MODELS=gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite,gemini-2.0-flash-001,gemini-2.0-flash-lite-001
+# 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_TITLE_MODEL=gemini-2.0-flash-lite-001

@@ -196,7 +175,7 @@ GOOGLE_KEY=user_provided
 #============#

 OPENAI_API_KEY=user_provided
-# OPENAI_MODELS=gpt-5,gpt-5-codex,gpt-5-mini,gpt-5-nano,o3-pro,o3,o4-mini,gpt-4.1,gpt-4.1-mini,gpt-4.1-nano,o3-mini,o1-pro,o1,gpt-4o,gpt-4o-mini
+# OPENAI_MODELS=o1,o1-mini,o1-preview,gpt-4o,gpt-4.5-preview,chatgpt-4o-latest,gpt-4o-mini,gpt-3.5-turbo-0125,gpt-3.5-turbo-0301,gpt-3.5-turbo,gpt-4,gpt-4-0613,gpt-4-vision-preview,gpt-3.5-turbo-0613,gpt-3.5-turbo-16k-0613,gpt-4-0125-preview,gpt-4-turbo-preview,gpt-4-1106-preview,gpt-3.5-turbo-1106,gpt-3.5-turbo-instruct,gpt-3.5-turbo-instruct-0914,gpt-3.5-turbo-16k

 DEBUG_OPENAI=false

@@ -370,11 +349,6 @@ 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
@@ -459,15 +433,10 @@ OPENID_CALLBACK_URL=/oauth/openid/callback
 OPENID_REQUIRED_ROLE=
 OPENID_REQUIRED_ROLE_TOKEN_KIND=
 OPENID_REQUIRED_ROLE_PARAMETER_PATH=
-OPENID_ADMIN_ROLE=
-OPENID_ADMIN_ROLE_PARAMETER_PATH=
-OPENID_ADMIN_ROLE_TOKEN_KIND=
 # Set to determine which user info property returned from OpenID Provider to store as the User's username
 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=
@@ -484,26 +453,11 @@ 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_USERINFO_REQUIRED=
-OPENID_ON_BEHALF_FLOW_USERINFO_SCOPE="user.read" # example for Scope Needed for Microsoft Graph API
+OPENID_ON_BEHALF_FLOW_FOR_USERINFRO_REQUIRED=
+OPENID_ON_BEHALF_FLOW_USERINFRO_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.
@@ -531,21 +485,6 @@ SAML_IMAGE_URL=
 # SAML_USE_AUTHN_RESPONSE_SIGNED=


-#===============================================#
-# Microsoft Graph API / Entra ID Integration  #
-#===============================================#
-
-# Enable Entra ID people search integration in permissions/sharing system
-# When enabled, the people picker will search both local database and Entra ID
-USE_ENTRA_ID_FOR_PEOPLE_SEARCH=false
-
-# When enabled, entra id groups owners will be considered as members of the group
-ENTRA_ID_INCLUDE_OWNERS_AS_MEMBERS=false
-
-# Microsoft Graph API scopes needed for people/group search
-# Default scopes provide access to user profiles and group memberships
-OPENID_GRAPH_SCOPES=User.Read,People.Read,GroupMember.Read.All
-
 # LDAP
 LDAP_URL=
 LDAP_BIND_DN=
@@ -576,18 +515,6 @@ EMAIL_PASSWORD=
 EMAIL_FROM_NAME=
 EMAIL_FROM=noreply@librechat.ai

-#========================#
-#      Mailgun API       #
-#========================#
-
-# MAILGUN_API_KEY=your-mailgun-api-key
-# MAILGUN_DOMAIN=mg.yourdomain.com
-# EMAIL_FROM=noreply@yourdomain.com
-# EMAIL_FROM_NAME="LibreChat"
-
-# # Optional: For EU region
-# MAILGUN_HOST=https://api.eu.mailgun.net
-
 #========================#
 # Firebase CDN           #
 #========================#
@@ -636,10 +563,6 @@ 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                         #
 #===================================================#
@@ -653,54 +576,15 @@ HELP_AND_FAQ_URL=https://librechat.ai
 # Google tag manager id
 #ANALYTICS_GTM_ID=user provided google tag manager id

-# limit conversation file imports to a certain number of bytes in size to avoid the container
-# maxing out memory limitations by unremarking this line and supplying a file size in bytes
-# such as the below example of 250 mib
-# CONVERSATION_IMPORT_MAX_FILE_SIZE_BYTES=262144000
-
-
 #===============#
 # REDIS Options #
 #===============#

-# Enable Redis for caching and session storage
+# REDIS_URI=10.10.10.10:6379
 # USE_REDIS=true

-# Single Redis instance
-# REDIS_URI=redis://127.0.0.1:6379
-
-# Redis cluster (multiple nodes)
-# REDIS_URI=redis://127.0.0.1:7001,redis://127.0.0.1:7002,redis://127.0.0.1:7003
-
-# Redis with TLS/SSL encryption and CA certificate
-# REDIS_URI=rediss://127.0.0.1:6380
-# REDIS_CA=/path/to/ca-cert.pem
-
-# Elasticache may need to use an alternate dnsLookup for TLS connections.  see "Special Note: Aws Elasticache Clusters with TLS" on this webpage: https://www.npmjs.com/package/ioredis
-# Enable alternative dnsLookup for redis
-# REDIS_USE_ALTERNATIVE_DNS_LOOKUP=true
-
-# Redis authentication (if required)
-# REDIS_USERNAME=your_redis_username
-# REDIS_PASSWORD=your_redis_password
-
-# Redis key prefix configuration
-# Use environment variable name for dynamic prefix (recommended for cloud deployments)
-# REDIS_KEY_PREFIX_VAR=K_REVISION
-# Or use static prefix directly
-# REDIS_KEY_PREFIX=librechat
-
-# Redis connection limits
-# REDIS_MAX_LISTENERS=40
-
-# Redis ping interval in seconds (0 = disabled, >0 = enabled)
-# When set to a positive integer, Redis clients will ping the server at this interval to keep connections alive
-# When unset or 0, no pinging is performed (recommended for most use cases)
-# REDIS_PING_INTERVAL=300
-
-# Force specific cache namespaces to use in-memory storage even when Redis is enabled
-# Comma-separated list of CacheKeys (e.g., ROLES,MESSAGES)
-# FORCED_IN_MEMORY_CACHE_NAMESPACES=ROLES,MESSAGES
+# USE_REDIS_CLUSTER=true
+# REDIS_CA=/path/to/ca.crt

 #==================================================#
 #                      Others                      #
@@ -761,17 +645,4 @@ OPENWEATHER_API_KEY=
 # Reranker (Required)
 # JINA_API_KEY=your_jina_api_key
 # or
-# COHERE_API_KEY=your_cohere_api_key
-
-#======================#
-# MCP Configuration    #
-#======================#
-
-# Treat 401/403 responses as OAuth requirement when no oauth metadata found
-# MCP_OAUTH_ON_AUTH_ERROR=true
-
-# Timeout for OAuth detection requests in milliseconds
-# MCP_OAUTH_DETECTION_TIMEOUT=5000
-
-# Cache connection status checks for this many milliseconds to avoid expensive verification
-# MCP_CONNECTION_CHECK_TTL=60000
+# COHERE_API_KEY=your_cohere_api_key
--- a/.github/CONTRIBUTING.md
+++ b/.github/CONTRIBUTING.md
@@ -30,8 +30,8 @@ Project maintainers have the right and responsibility to remove, edit, or reject
 2. Install typescript globally: `npm i -g typescript`.
 3. Run `npm ci` to install dependencies.
 4. Build the data provider: `npm run build:data-provider`.
-5. Build data schemas: `npm run build:data-schemas`.
-6. Build API methods: `npm run build:api`.
+5. Build MCP: `npm run build:mcp`.
+6. Build data schemas: `npm run build:data-schemas`.
 7. Setup and run unit tests:
    - Copy `.env.test`: `cp api/test/.env.test.example api/test/.env.test`.
    - Run backend unit tests: `npm run test:api`.
@@ -147,7 +147,7 @@ Apply the following naming conventions to branches, labels, and other Git-relate
 ## 8. Module Import Conventions

 - `npm` packages first, 
-     - from longest line (top) to shortest (bottom)
+     - from shortest line (top) to longest (bottom)

 - Followed by typescript types (pertains to data-provider and client workspaces)
     - longest line (top) to shortest (bottom)
@@ -157,8 +157,6 @@ Apply the following naming conventions to branches, labels, and other Git-relate
     - longest line (top) to shortest (bottom)
     - imports with alias `~` treated the same as relative import with respect to line length

-**Note:** ESLint will automatically enforce these import conventions when you run `npm run lint --fix` or through pre-commit hooks.
-
 ---

 Please ensure that you adapt this summary to fit the specific context and nuances of your project.
--- a/.github/workflows/backend-review.yml
+++ b/.github/workflows/backend-review.yml
@@ -7,7 +7,6 @@ on:
      - release/*
    paths:
      - 'api/**'
-      - 'packages/**'
 jobs:
  tests_Backend:
    name: Run Backend unit tests
@@ -37,12 +36,12 @@ jobs:
      - name: Install Data Provider Package
        run: npm run build:data-provider

+      - name: Install MCP Package
+        run: npm run build:mcp
+
      - name: Install Data Schemas Package
        run: npm run build:data-schemas

-      - name: Install API Package
-        run: npm run build:api
-
      - name: Create empty auth.json file
        run: |
          mkdir -p api/data
@@ -67,8 +66,5 @@ jobs:
      - name: Run librechat-data-provider unit tests
        run: cd packages/data-provider && npm run test:ci

-      - name: Run @librechat/data-schemas unit tests
-        run: cd packages/data-schemas && npm run test:ci
-
-      - name: Run @librechat/api unit tests
-        run: cd packages/api && npm run test:ci
+      - name: Run librechat-mcp unit tests
+        run: cd packages/mcp && npm run test:ci
--- a/.github/workflows/cache-integration-tests.yml
+++ b/.github/workflows/cache-integration-tests.yml
@@ -1,78 +0,0 @@
-name: Cache Integration Tests
-
-on:
-  pull_request:
-    branches:
-      - main
-      - dev
-      - release/*
-    paths:
-      - 'packages/api/src/cache/**'
-      - 'redis-config/**'
-      - '.github/workflows/cache-integration-tests.yml'
-
-jobs:
-  cache_integration_tests:
-    name: Run Cache Integration Tests
-    timeout-minutes: 30
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Use Node.js 20.x
-        uses: actions/setup-node@v4
-        with:
-          node-version: 20
-          cache: 'npm'
-
-      - name: Install Redis tools
-        run: |
-          sudo apt-get update
-          sudo apt-get install -y redis-server redis-tools
-
-      - name: Start Single Redis Instance
-        run: |
-          redis-server --daemonize yes --port 6379
-          sleep 2
-          # Verify single Redis is running
-          redis-cli -p 6379 ping || exit 1
-
-      - name: Start Redis Cluster
-        working-directory: redis-config
-        run: |
-          chmod +x start-cluster.sh stop-cluster.sh
-          ./start-cluster.sh
-          sleep 10
-          # Verify cluster is running
-          redis-cli -p 7001 cluster info || exit 1
-          redis-cli -p 7002 cluster info || exit 1
-          redis-cli -p 7003 cluster info || exit 1
-
-      - name: Install dependencies
-        run: npm ci
-
-      - name: Build packages
-        run: |
-          npm run build:data-provider
-          npm run build:data-schemas
-          npm run build:api
-
-      - name: Run cache integration tests
-        working-directory: packages/api
-        env:
-          NODE_ENV: test
-          USE_REDIS: true
-          REDIS_URI: redis://127.0.0.1:6379
-          REDIS_CLUSTER_URI: redis://127.0.0.1:7001,redis://127.0.0.1:7002,redis://127.0.0.1:7003
-        run: npm run test:cache:integration
-
-      - name: Stop Redis Cluster
-        if: always()
-        working-directory: redis-config
-        run: ./stop-cluster.sh || true
-
-      - name: Stop Single Redis Instance
-        if: always()
-        run: redis-cli -p 6379 shutdown || true
--- a/.github/workflows/client.yml
+++ b/.github/workflows/client.yml
@@ -1,58 +0,0 @@
-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
--- a/.github/workflows/data-provider.yml
+++ b/.github/workflows/data-provider.yml
@@ -1,4 +1,4 @@
-name: Publish `librechat-data-provider` to NPM
+name: Node.js Package

 on:
  push:
@@ -6,12 +6,6 @@ on:
      - main
    paths:
      - 'packages/data-provider/package.json'
-  workflow_dispatch:
-    inputs:
-      reason:
-        description: 'Reason for manual trigger'
-        required: false
-        default: 'Manual publish requested'

 jobs:
  build:
@@ -20,7 +14,7 @@ jobs:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
-          node-version: 20
+          node-version: 16
      - run: cd packages/data-provider && npm ci
      - run: cd packages/data-provider && npm run build

@@ -31,7 +25,7 @@ jobs:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
-          node-version: 20
+          node-version: 16
          registry-url: 'https://registry.npmjs.org'
      - run: cd packages/data-provider && npm ci
      - run: cd packages/data-provider && npm run build
--- a/.github/workflows/data-schemas.yml
+++ b/.github/workflows/data-schemas.yml
@@ -22,7 +22,7 @@ jobs:
      - name: Use Node.js
        uses: actions/setup-node@v4
        with:
-          node-version: '20.x'
+          node-version: '18.x'
          
      - name: Install dependencies
        run: cd packages/data-schemas && npm ci
--- a/.github/workflows/deploy-dev.yml
+++ b/.github/workflows/deploy-dev.yml
@@ -2,7 +2,7 @@ name: Update Test Server

 on:
  workflow_run:
-    workflows: ["Docker Dev Branch Images Build"]
+    workflows: ["Docker Dev Images Build"]
    types:
      - completed
  workflow_dispatch:
@@ -12,8 +12,7 @@ jobs:
    runs-on: ubuntu-latest
    if: |
      github.repository == 'danny-avila/LibreChat' &&
-      (github.event_name == 'workflow_dispatch' || 
-       (github.event.workflow_run.conclusion == 'success' && github.event.workflow_run.head_branch == 'dev'))
+      (github.event_name == 'workflow_dispatch' || github.event.workflow_run.conclusion == 'success')
    steps:
    - name: Checkout repository
      uses: actions/checkout@v4
@@ -30,17 +29,13 @@ jobs:
        DO_USER: ${{ secrets.DO_USER }}
      run: |
        ssh -o StrictHostKeyChecking=no ${DO_USER}@${DO_HOST} << EOF
-        sudo -i -u danny bash << 'EEOF'
+        sudo -i -u danny bash << EEOF
        cd ~/LibreChat && \
        git fetch origin main && \
-        sudo npm run stop:deployed && \
-        sudo docker images --format "{{.Repository}}:{{.ID}}" | grep -E "lc-dev|librechat" | cut -d: -f2 | xargs -r sudo docker rmi -f || true && \
-        sudo npm run update:deployed && \
-        git checkout dev && \
-        git pull origin dev && \
+        npm run update:deployed && \
        git checkout do-deploy && \
-        git rebase dev && \
-        sudo npm run start:deployed && \
+        git rebase main && \
+        npm run start:deployed && \
        echo "Update completed. Application should be running now."
        EEOF
        EOF
--- a/.github/workflows/dev-branch-images.yml
+++ b/.github/workflows/dev-branch-images.yml
@@ -1,72 +0,0 @@
-name: Docker Dev Branch Images Build
-
-on:
-  workflow_dispatch:
-  push:
-    branches:
-      - dev
-    paths:
-      - 'api/**'
-      - 'client/**'
-      - 'packages/**'
-
-jobs:
-  build:
-    runs-on: ubuntu-latest
-    strategy:
-      matrix:
-        include:
-          - target: api-build
-            file: Dockerfile.multi
-            image_name: lc-dev-api
-          - target: node
-            file: Dockerfile
-            image_name: lc-dev
-
-    steps:
-      # Check out the repository
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      # Set up QEMU
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
-
-      # Set up Docker Buildx
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      # Log in to GitHub Container Registry
-      - name: Log in to GitHub Container Registry
-        uses: docker/login-action@v2
-        with:
-          registry: ghcr.io
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-
-      # Login to Docker Hub
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-
-      # Prepare the environment
-      - name: Prepare environment
-        run: |
-          cp .env.example .env
-
-      # Build and push Docker images for each target
-      - name: Build and push Docker images
-        uses: docker/build-push-action@v5
-        with:
-          context: .
-          file: ${{ matrix.file }}
-          push: true
-          tags: |
-            ghcr.io/${{ github.repository_owner }}/${{ matrix.image_name }}:${{ github.sha }}
-            ghcr.io/${{ github.repository_owner }}/${{ matrix.image_name }}:latest
-            ${{ secrets.DOCKERHUB_USERNAME }}/${{ matrix.image_name }}:${{ github.sha }}
-            ${{ secrets.DOCKERHUB_USERNAME }}/${{ matrix.image_name }}:latest
-          platforms: linux/amd64,linux/arm64
-          target: ${{ matrix.target }} 
--- a/.github/workflows/frontend-review.yml
+++ b/.github/workflows/frontend-review.yml
@@ -8,7 +8,7 @@ on:
      - release/*
    paths:
      - 'client/**'
-      - 'packages/data-provider/**'
+      - 'packages/**'

 jobs:
  tests_frontend_ubuntu:
--- a/.github/workflows/generate-release-changelog-pr.yml
+++ b/.github/workflows/generate-release-changelog-pr.yml
@@ -0,0 +1,95 @@
+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 ${{ github.ref_name }} above previous releases.
--- a/.github/workflows/generate-unreleased-changelog-pr.yml
+++ b/.github/workflows/generate-unreleased-changelog-pr.yml
@@ -0,0 +1,107 @@
+name: Generate Unreleased Changelog PR
+
+on:
+  schedule:
+    - cron: "0 0 * * 1"  # Runs every Monday at 00:00 UTC
+  workflow_dispatch:
+
+jobs:
+  generate-unreleased-changelog-pr:
+    permissions:
+      contents: write    # Needed for pushing commits and creating branches.
+      pull-requests: write
+    runs-on: ubuntu-latest
+    steps:
+      # 1. Checkout the repository on main.
+      - name: Checkout Repository on Main
+        uses: actions/checkout@v4
+        with:
+          ref: main
+          fetch-depth: 0
+
+      # 4. Get the latest version tag.
+      - name: Get Latest Tag
+        id: get_latest_tag
+        run: |
+          LATEST_TAG=$(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.
--- a/.github/workflows/helmcharts.yml
+++ b/.github/workflows/helmcharts.yml
@@ -4,13 +4,12 @@ name: Build Helm Charts on Tag
 on:
  push:
    tags:
-      - "chart-*"
+      - "*"

 jobs:
  release:
    permissions:
      contents: write
-      packages: write
    runs-on: ubuntu-latest
    steps:
      - name: Checkout
@@ -27,49 +26,15 @@ jobs:
        uses: azure/setup-helm@v4
        env:
          GITHUB_TOKEN: "${{ secrets.GITHUB_TOKEN }}"
-      
      - name: Build Subchart Deps
        run: |
-          cd helm/librechat
-          helm dependency build  
-          cd ../librechat-rag-api
-          helm dependency build      
+          cd helm/librechat-rag-api
+          helm dependency build 

-      - name: Get Chart Version
-        id: chart-version
-        run: |
-          CHART_VERSION=$(echo "${{ github.ref_name }}" | cut -d'-' -f2)
-          echo "CHART_VERSION=${CHART_VERSION}" >> "$GITHUB_OUTPUT"
-
-      # Log in to GitHub Container Registry
-      - name: Log in to GitHub Container Registry
-        uses: docker/login-action@v3
+      - name: Run chart-releaser
+        uses: helm/chart-releaser-action@v1.6.0
        with:
-          registry: ghcr.io
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-
-      # Run Helm OCI Charts Releaser
-      # This is for the librechat chart
-      - name: Release Helm OCI Charts for librechat
-        uses: appany/helm-oci-chart-releaser@v0.4.2
-        with:
-          name: librechat
-          repository: ${{ github.actor }}/librechat-chart
-          tag: ${{ steps.chart-version.outputs.CHART_VERSION }}
-          path: helm/librechat
-          registry: ghcr.io
-          registry_username: ${{ github.actor }}
-          registry_password: ${{ secrets.GITHUB_TOKEN }}
-      
-      # this is for the librechat-rag-api chart
-      - name: Release Helm OCI Charts for librechat-rag-api
-        uses: appany/helm-oci-chart-releaser@v0.4.2
-        with:
-          name: librechat-rag-api
-          repository: ${{ github.actor }}/librechat-chart
-          tag: ${{ steps.chart-version.outputs.CHART_VERSION }}
-          path: helm/librechat-rag-api
-          registry: ghcr.io
-          registry_username: ${{ github.actor }}
-          registry_password: ${{ secrets.GITHUB_TOKEN }}
+          charts_dir: helm
+          skip_existing: true
+        env:
+          CR_TOKEN: "${{ secrets.GITHUB_TOKEN }}"
--- a/.github/workflows/i18n-unused-keys.yml
+++ b/.github/workflows/i18n-unused-keys.yml
@@ -1,24 +1,16 @@
 name: Detect Unused i18next Strings

-# This workflow checks for unused i18n keys in translation files.
-# It has special handling for:
-# - com_ui_special_var_* keys that are dynamically constructed
-# - com_agents_category_* keys that are stored in the database and used dynamically
-
 on:
  pull_request:
    paths:
      - "client/src/**"
      - "api/**"
-      - "packages/data-provider/src/**"
-      - "packages/client/**"
-      - "packages/data-schemas/src/**"

 jobs:
  detect-unused-i18n-keys:
    runs-on: ubuntu-latest
    permissions:
-      pull-requests: write
+      pull-requests: write  # Required for posting PR comments
    steps:
      - name: Checkout repository
        uses: actions/checkout@v3
@@ -30,7 +22,7 @@ jobs:

          # Define paths
          I18N_FILE="client/src/locales/en/translation.json"
-          SOURCE_DIRS=("client/src" "api" "packages/data-provider/src" "packages/client" "packages/data-schemas/src")
+          SOURCE_DIRS=("client/src" "api" "packages/data-provider/src")

          # Check if translation file exists
          if [[ ! -f "$I18N_FILE" ]]; then
@@ -58,31 +50,6 @@ jobs:
                fi
              done
              
-              # Also check if the key is directly used somewhere
-              if [[ "$FOUND" == false ]]; then
-                for DIR in "${SOURCE_DIRS[@]}"; do
-                  if grep -r --include=\*.{js,jsx,ts,tsx} -q "$KEY" "$DIR"; then
-                    FOUND=true
-                    break
-                  fi
-                done
-              fi
-            # Special case for agent category keys that are dynamically used from database
-            elif [[ "$KEY" == com_agents_category_* ]]; then
-              # Check if agent category localization is being used
-              for DIR in "${SOURCE_DIRS[@]}"; do
-                # Check for dynamic category label/description usage
-                if grep -r --include=\*.{js,jsx,ts,tsx} -E "category\.(label|description).*startsWith.*['\"]com_" "$DIR" > /dev/null 2>&1 || \
-                   # Check for the method that defines these keys
-                   grep -r --include=\*.{js,jsx,ts,tsx} "ensureDefaultCategories" "$DIR" > /dev/null 2>&1 || \
-                   # Check for direct usage in agentCategory.ts
-                   grep -r --include=\*.ts -E "label:.*['\"]$KEY['\"]" "$DIR" > /dev/null 2>&1 || \
-                   grep -r --include=\*.ts -E "description:.*['\"]$KEY['\"]" "$DIR" > /dev/null 2>&1; then
-                  FOUND=true
-                  break
-                fi
-              done
-              
              # Also check if the key is directly used somewhere
              if [[ "$FOUND" == false ]]; then
                for DIR in "${SOURCE_DIRS[@]}"; do
--- a/.github/workflows/locize-i18n-sync.yml
+++ b/.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@v2
+        uses: locize/download@v1
        with:
          project-id: ${{ secrets.LOCIZE_PROJECT_ID }}
          path: "client/src/locales"
--- a/.github/workflows/unused-packages.yml
+++ b/.github/workflows/unused-packages.yml
@@ -7,7 +7,6 @@ on:
      - 'package-lock.json'
      - 'client/**'
      - 'api/**'
-      - 'packages/client/**'

 jobs:
  detect-unused-packages:
@@ -29,7 +28,7 @@ jobs:

      - name: Validate JSON files
        run: |
-          for FILE in package.json client/package.json api/package.json packages/client/package.json; do
+          for FILE in package.json client/package.json api/package.json; do
            if [[ -f "$FILE" ]]; then
              jq empty "$FILE" || (echo "::error title=Invalid JSON::$FILE is invalid" && exit 1)
            fi
@@ -64,31 +63,12 @@ jobs:
            local folder=$1
            local output_file=$2
            if [[ -d "$folder" ]]; then
-              # Extract require() statements
-              grep -rEho "require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
+              grep -rEho "require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)" "$folder" --include=\*.{js,ts,mjs,cjs} | \
                sed -E "s/require\\(['\"]([a-zA-Z0-9@/._-]+)['\"]\\)/\1/" > "$output_file"
          
-              # Extract ES6 imports - various patterns
-              # import x from 'module'
-              grep -rEho "import .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,tsx,jsx,mjs,cjs} | \
+              grep -rEho "import .* from ['\"]([a-zA-Z0-9@/._-]+)['\"]" "$folder" --include=\*.{js,ts,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"
@@ -98,80 +78,13 @@ 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=$(depcheck --json | jq -r '.dependencies | join("\n")' || echo "")
-            # Exclude dependencies used in scripts, code, and workspace packages
-            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat root_used_deps.txt root_used_code.txt root_workspace_deps.txt | sort) || echo "")
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat root_used_deps.txt root_used_code.txt | sort) || echo "")
            echo "ROOT_UNUSED<<EOF" >> $GITHUB_ENV
            echo "$UNUSED" >> $GITHUB_ENV
            echo "EOF" >> $GITHUB_ENV
@@ -184,10 +97,7 @@ jobs:
            chmod -R 755 client
            cd client
            UNUSED=$(depcheck --json | jq -r '.dependencies | join("\n")' || 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$" || echo "")
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../client_used_deps.txt ../client_used_code.txt | sort) || echo "")
            echo "CLIENT_UNUSED<<EOF" >> $GITHUB_ENV
            echo "$UNUSED" >> $GITHUB_ENV
            echo "EOF" >> $GITHUB_ENV
@@ -201,8 +111,7 @@ jobs:
            chmod -R 755 api
            cd api
            UNUSED=$(depcheck --json | jq -r '.dependencies | join("\n")' || 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 "")
+            UNUSED=$(comm -23 <(echo "$UNUSED" | sort) <(cat ../api_used_deps.txt ../api_used_code.txt | sort) || echo "")
            echo "API_UNUSED<<EOF" >> $GITHUB_ENV
            echo "$UNUSED" >> $GITHUB_ENV
            echo "EOF" >> $GITHUB_ENV
--- a/.gitignore
+++ b/.gitignore
@@ -13,9 +13,6 @@ pids
 *.seed
 .git

-# CI/CD data
-test-image*
-
 # Directory for instrumented libs generated by jscoverage/JSCover
 lib-cov

@@ -58,7 +55,6 @@ bower_components/
 # AI
 .clineignore
 .cursor
-.aider*

 # Floobits
 .floo
@@ -128,13 +124,3 @@ helm/**/.values.yaml

 # SAML Idp cert
 *.cert
-
-# AI Assistants
-/.claude/
-/.cursor/
-/.copilot/
-/.aider/
-/.openai/
-/.tabnine/
-/.codeium
-*.local.md
--- a/.husky/pre-commit
+++ b/.husky/pre-commit
@@ -1,2 +1,5 @@
+#!/usr/bin/env sh 
+set -e
+. "$(dirname -- "$0")/_/husky.sh"
 [ -n "$CI" ] && exit 0
 npx lint-staged --config ./.husky/lint-staged.config.js
--- a/.vscode/launch.json
+++ b/.vscode/launch.json
@@ -8,8 +8,7 @@
      "skipFiles": ["<node_internals>/**"],
      "program": "${workspaceFolder}/api/server/index.js",
      "env": {
-        "NODE_ENV": "production",
-        "NODE_TLS_REJECT_UNAUTHORIZED": "0"
+        "NODE_ENV": "production"
      },
      "console": "integratedTerminal",
      "envFile": "${workspaceFolder}/.env"
--- a/21
+++ b/21
@@ -1,4 +1,4 @@
-# v0.8.1-rc1
+# v0.7.8

 # Base node image
 FROM node:20-alpine AS node
@@ -19,31 +19,24 @@ WORKDIR /app

 USER node

-COPY --chown=node:node package.json package-lock.json ./
-COPY --chown=node:node api/package.json ./api/package.json
-COPY --chown=node:node client/package.json ./client/package.json
-COPY --chown=node:node packages/data-provider/package.json ./packages/data-provider/package.json
-COPY --chown=node:node packages/data-schemas/package.json ./packages/data-schemas/package.json
-COPY --chown=node:node packages/api/package.json ./packages/api/package.json
+COPY --chown=node:node . .

 RUN \
    # Allow mounting of these files, which have no default
    touch .env ; \
    # Create directories for the volumes to inherit the correct permissions
-    mkdir -p /app/client/public/images /app/api/logs /app/uploads ; \
+    mkdir -p /app/client/public/images /app/api/logs ; \
    npm config set fetch-retry-maxtimeout 600000 ; \
    npm config set fetch-retries 5 ; \
    npm config set fetch-retry-mintimeout 15000 ; \
-    npm ci --no-audit
-
-COPY --chown=node:node . .
-
-RUN \
+    npm install --no-audit; \
    # React client build
    NODE_OPTIONS="--max-old-space-size=2048" npm run frontend; \
    npm prune --production; \
    npm cache clean --force

+RUN mkdir -p /app/client/public/images /app/api/logs
+
 # Node API setup
 EXPOSE 3080
 ENV HOST=0.0.0.0
@@ -54,4 +47,4 @@ CMD ["npm", "run", "backend"]
 # WORKDIR /usr/share/nginx/html
 # COPY --from=node /app/client/dist /usr/share/nginx/html
 # COPY client/nginx.conf /etc/nginx/conf.d/default.conf
-# ENTRYPOINT ["nginx", "-g", "daemon off;"]
+# ENTRYPOINT ["nginx", "-g", "daemon off;"]
--- a/Dockerfile.multi
+++ b/Dockerfile.multi
@@ -1,5 +1,5 @@
 # Dockerfile.multi
-# v0.8.1-rc1
+# v0.7.8

 # Base for all builds
 FROM node:20-alpine AS base-min
@@ -14,9 +14,8 @@ RUN npm config set fetch-retry-maxtimeout 600000 && \
    npm config set fetch-retry-mintimeout 15000
 COPY package*.json ./
 COPY packages/data-provider/package*.json ./packages/data-provider/
-COPY packages/api/package*.json ./packages/api/
+COPY packages/mcp/package*.json ./packages/mcp/
 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/

@@ -25,40 +24,31 @@ FROM base-min AS base
 WORKDIR /app
 RUN npm ci

-# Build `data-provider` package
+# Build data-provider
 FROM base AS data-provider-build
 WORKDIR /app/packages/data-provider
 COPY packages/data-provider ./
 RUN npm run build

-# Build `data-schemas` package
+# Build mcp package
+FROM base AS mcp-build
+WORKDIR /app/packages/mcp
+COPY packages/mcp ./
+COPY --from=data-provider-build /app/packages/data-provider/dist /app/packages/data-provider/dist
+RUN npm run build
+
+# Build data-schemas
 FROM base AS data-schemas-build
 WORKDIR /app/packages/data-schemas
 COPY packages/data-schemas ./
 COPY --from=data-provider-build /app/packages/data-provider/dist /app/packages/data-provider/dist
 RUN npm run build

-# Build `api` package
-FROM base AS api-package-build
-WORKDIR /app/packages/api
-COPY packages/api ./
-COPY --from=data-provider-build /app/packages/data-provider/dist /app/packages/data-provider/dist
-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

@@ -73,8 +63,8 @@ RUN npm ci --omit=dev
 COPY api ./api
 COPY config ./config
 COPY --from=data-provider-build /app/packages/data-provider/dist ./packages/data-provider/dist
+COPY --from=mcp-build /app/packages/mcp/dist ./packages/mcp/dist
 COPY --from=data-schemas-build /app/packages/data-schemas/dist ./packages/data-schemas/dist
-COPY --from=api-package-build /app/packages/api/dist ./packages/api/dist
 COPY --from=client-build /app/client/dist ./client/dist
 WORKDIR /app/api
 EXPOSE 3080
--- a/README.md
+++ b/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 Responses API (incl. Azure)
+  - Anthropic (Claude), AWS Bedrock, OpenAI, Azure OpenAI, Google, Vertex AI, OpenAI Assistants API (incl. Azure)
  - [Custom Endpoints](https://www.librechat.ai/docs/quick_start/custom_endpoints): Use any OpenAI-compatible API with LibreChat, no proxy required
  - Compatible with [Local & Remote AI Providers](https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints):
    - Ollama, groq, Cohere, Mistral AI, Apple MLX, koboldcpp, together.ai,
@@ -65,17 +65,15 @@

 - 🔦 **Agents & Tools Integration**:  
  - **[LibreChat Agents](https://www.librechat.ai/docs/features/agents)**:
-    - No-Code Custom Assistants: Build specialized, AI-driven helpers
-    - Agent Marketplace: Discover and deploy community-built agents
-    - Collaborative Sharing: Share agents with specific users and groups
-    - Flexible & Extensible: Use MCP Servers, tools, file search, code execution, and more
-    - Compatible with Custom Endpoints, OpenAI, Azure, Anthropic, AWS Bedrock, Google, Vertex AI, Responses API, and more
+    - 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
    - [Model Context Protocol (MCP) Support](https://modelcontextprotocol.io/clients#librechat) for Tools
+  - Use LibreChat Agents and OpenAI Assistants with Files, Code Interpreter, Tools, and API Actions

 - 🔍 **Web Search**:  
  - Search the internet and retrieve relevant information to enhance your AI context
  - Combines search providers, content scrapers, and result rerankers for optimal results
-  - **Customizable Jina Reranking**: Configure custom Jina API URLs for reranking services
  - **[Learn More →](https://www.librechat.ai/docs/features/web_search)**

 - 🪄 **Generative UI with Code Artifacts**:  
@@ -90,18 +88,15 @@
  - Create, Save, & Share Custom Presets  
  - Switch between AI Endpoints and Presets mid-chat
  - Edit, Resubmit, and Continue Messages with Conversation branching  
-  - Create and share prompts with specific users and groups
  - [Fork Messages & Conversations](https://www.librechat.ai/docs/features/fork) for Advanced Context control

 - 💬 **Multimodal & File Interactions**:  
  - Upload and analyze images with Claude 3, GPT-4.5, GPT-4o, o1, Llama-Vision, and Gemini 📸  
  - Chat with Files using Custom Endpoints, OpenAI, Azure, Anthropic, AWS Bedrock, & Google 🗃️

- 🌎 **Multilingual UI**:
-  - English, 中文 (简体), 中文 (繁體), العربية, Deutsch, Español, Français, Italiano
-  - Polski, Português (PT), Português (BR), Русский, 日本語, Svenska, 한국어, Tiếng Việt
-  - Türkçe, Nederlands, עברית, Català, Čeština, Dansk, Eesti, فارسی
-  - Suomi, Magyar, Հայերեն, Bahasa Indonesia, ქართული, Latviešu, ไทย, ئۇيغۇرچە
+- 🌎 **Multilingual UI**:  
+  - English, 中文, Deutsch, Español, Français, Italiano, Polski, Português Brasileiro
+  - Русский, 日本語, Svenska, 한국어, Tiếng Việt, 繁體中文, العربية, Türkçe, Nederlands, עברית

 - 🧠 **Reasoning UI**:  
  - Dynamic Reasoning UI for Chain-of-Thought/Reasoning AI models like DeepSeek-R1
@@ -155,8 +150,8 @@ Click on the thumbnail to open the video☝️

 **Other:**
  - **Website:** [librechat.ai](https://librechat.ai)
-  - **Documentation:** [librechat.ai/docs](https://librechat.ai/docs)
-  - **Blog:** [librechat.ai/blog](https://librechat.ai/blog)
+  - **Documentation:** [docs.librechat.ai](https://docs.librechat.ai)
+  - **Blog:** [blog.librechat.ai](https://blog.librechat.ai)

 ---

--- a/api/app/clients/AnthropicClient.js
+++ b/api/app/clients/AnthropicClient.js
@@ -1,5 +1,4 @@
 const Anthropic = require('@anthropic-ai/sdk');
-const { logger } = require('@librechat/data-schemas');
 const { HttpsProxyAgent } = require('https-proxy-agent');
 const {
  Constants,
@@ -10,18 +9,7 @@ const {
  getResponseSender,
  validateVisionModel,
 } = require('librechat-data-provider');
-const { sleep, SplitStreamHandler: _Handler } = require('@librechat/agents');
-const {
-  Tokenizer,
-  createFetch,
-  matchModelName,
-  getClaudeHeaders,
-  getModelMaxTokens,
-  configureReasoning,
-  checkPromptCacheSupport,
-  getModelMaxOutputTokens,
-  createStreamEventHandlers,
-} = require('@librechat/api');
+const { SplitStreamHandler: _Handler } = require('@librechat/agents');
 const {
  truncateText,
  formatMessage,
@@ -30,9 +18,19 @@ const {
  parseParamFromPrompt,
  createContextHandlers,
 } = require('./prompts');
+const {
+  getClaudeHeaders,
+  configureReasoning,
+  checkPromptCacheSupport,
+} = require('~/server/services/Endpoints/anthropic/helpers');
+const { getModelMaxTokens, getModelMaxOutputTokens, matchModelName } = require('~/utils');
 const { spendTokens, spendStructuredTokens } = require('~/models/spendTokens');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
+const { createFetch, createStreamEventHandlers } = require('./generators');
+const Tokenizer = require('~/server/services/Tokenizer');
+const { sleep } = require('~/server/utils');
 const BaseClient = require('./BaseClient');
+const { logger } = require('~/config');

 const HUMAN_PROMPT = '\n\nHuman:';
 const AI_PROMPT = '\n\nAssistant:';
@@ -193,11 +191,10 @@ class AnthropicClient extends BaseClient {
        reverseProxyUrl: this.options.reverseProxyUrl,
      }),
      apiKey: this.apiKey,
-      fetchOptions: {},
    };

    if (this.options.proxy) {
-      options.fetchOptions.agent = new HttpsProxyAgent(this.options.proxy);
+      options.httpAgent = new HttpsProxyAgent(this.options.proxy);
    }

    if (this.options.reverseProxyUrl) {
--- a/api/app/clients/BaseClient.js
+++ b/api/app/clients/BaseClient.js
@@ -1,31 +1,22 @@
 const crypto = require('crypto');
 const fetch = require('node-fetch');
-const { logger } = require('@librechat/data-schemas');
 const {
-  getBalanceConfig,
-  extractFileContext,
-  encodeAndFormatAudios,
-  encodeAndFormatVideos,
-  encodeAndFormatDocuments,
-} = require('@librechat/api');
-const {
-  Constants,
-  ErrorTypes,
-  FileSources,
+  supportsBalanceCheck,
+  isAgentsEndpoint,
+  isParamEndpoint,
+  EModelEndpoint,
  ContentTypes,
  excludedKeys,
-  EModelEndpoint,
-  isParamEndpoint,
-  isAgentsEndpoint,
-  supportsBalanceCheck,
+  ErrorTypes,
+  Constants,
 } = require('librechat-data-provider');
 const { getMessages, saveMessage, updateMessage, saveConvo, getConvo } = require('~/models');
-const { getStrategyFunctions } = require('~/server/services/Files/strategies');
 const { checkBalance } = require('~/models/balanceMethods');
 const { truncateToolCallOutputs } = require('./prompts');
-const countTokens = require('~/server/utils/countTokens');
+const { addSpaceIfNeeded } = require('~/server/utils');
 const { getFiles } = require('~/models/File');
 const TextStream = require('./TextStream');
+const { logger } = require('~/config');

 class BaseClient {
  constructor(apiKey, options = {}) {
@@ -47,8 +38,6 @@ class BaseClient {
    this.conversationId;
    /** @type {string} */
    this.responseMessageId;
-    /** @type {string} */
-    this.parentMessageId;
    /** @type {TAttachment[]} */
    this.attachments;
    /** The key for the usage object's input tokens
@@ -120,17 +109,12 @@ class BaseClient {
  /**
   * Abstract method to record token usage. Subclasses must implement this method.
   * If a correction to the token usage is needed, the method should return an object with the corrected token counts.
-   * Should only be used if `recordCollectedUsage` was not used instead.
-   * @param {string} [model]
-   * @param {AppConfig['balance']} [balance]
   * @param {number} promptTokens
   * @param {number} completionTokens
   * @returns {Promise<void>}
   */
-  async recordTokenUsage({ model, balance, promptTokens, completionTokens }) {
+  async recordTokenUsage({ promptTokens, completionTokens }) {
    logger.debug('[BaseClient] `recordTokenUsage` not implemented.', {
-      model,
-      balance,
      promptTokens,
      completionTokens,
    });
@@ -199,8 +183,7 @@ class BaseClient {
    this.user = user;
    const saveOptions = this.getSaveOptions();
    this.abortController = opts.abortController ?? new AbortController();
-    const requestConvoId = overrideConvoId ?? opts.conversationId;
-    const conversationId = requestConvoId ?? crypto.randomUUID();
+    const conversationId = overrideConvoId ?? opts.conversationId ?? crypto.randomUUID();
    const parentMessageId = opts.parentMessageId ?? Constants.NO_PARENT;
    const userMessageId =
      overrideUserMessageId ?? opts.overrideParentMessageId ?? crypto.randomUUID();
@@ -215,22 +198,17 @@ class BaseClient {
      this.currentMessages[this.currentMessages.length - 1].messageId = head;
    }

-    if (opts.isRegenerate && responseMessageId.endsWith('_')) {
-      responseMessageId = crypto.randomUUID();
-    }
-
    this.responseMessageId = responseMessageId;

    return {
      ...opts,
      user,
      head,
-      saveOptions,
-      userMessageId,
-      requestConvoId,
      conversationId,
      parentMessageId,
+      userMessageId,
      responseMessageId,
+      saveOptions,
    };
  }

@@ -249,12 +227,11 @@ class BaseClient {
    const {
      user,
      head,
-      saveOptions,
-      userMessageId,
-      requestConvoId,
      conversationId,
      parentMessageId,
+      userMessageId,
      responseMessageId,
+      saveOptions,
    } = await this.setMessageOptions(opts);

    const userMessage = opts.isEdited
@@ -276,8 +253,7 @@ class BaseClient {
    }

    if (typeof opts?.onStart === 'function') {
-      const isNewConvo = !requestConvoId && parentMessageId === Constants.NO_PARENT;
-      opts.onStart(userMessage, responseMessageId, isNewConvo);
+      opts.onStart(userMessage, responseMessageId);
    }

    return {
@@ -583,7 +559,6 @@ class BaseClient {
  }

  async sendMessage(message, opts = {}) {
-    const appConfig = this.options.req?.config;
    /** @type {Promise<TMessage>} */
    let userMessagePromise;
    const { user, head, isEdited, conversationId, responseMessageId, saveOptions, userMessage } =
@@ -597,7 +572,7 @@ class BaseClient {
      });
    }

-    const { editedContent } = opts;
+    const { generation = '' } = opts;

    // It's not necessary to push to currentMessages
    // depending on subclass implementation of handling messages
@@ -612,40 +587,26 @@ class BaseClient {
          isCreatedByUser: false,
          model: this.modelOptions?.model ?? this.model,
          sender: this.sender,
+          text: generation,
        };
        this.currentMessages.push(userMessage, latestMessage);
-      } 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;
-            }
-          }
-        }
+      } else {
+        latestMessage.text = generation;
      }
      this.continued = true;
    } else {
      this.currentMessages.push(userMessage);
    }

-    /**
-     * When the userMessage is pushed to currentMessages, the parentMessage is the userMessageId.
-     * this only matters when buildMessages is utilizing the parentMessageId, and may vary on implementation
-     */
-    const parentMessageId = isEdited ? head : userMessage.messageId;
-    this.parentMessageId = parentMessageId;
    let {
      prompt: payload,
      tokenCountMap,
      promptTokens,
    } = await this.buildMessages(
      this.currentMessages,
-      parentMessageId,
+      // When the userMessage is pushed to currentMessages, the parentMessage is the userMessageId.
+      // this only matters when buildMessages is utilizing the parentMessageId, and may vary on implementation
+      isEdited ? head : userMessage.messageId,
      this.getBuildMessagesOptions(opts),
      opts,
    );
@@ -670,9 +631,9 @@ class BaseClient {
      }
    }

-    const balanceConfig = getBalanceConfig(appConfig);
+    const balance = this.options.req?.app?.locals?.balance;
    if (
-      balanceConfig?.enabled &&
+      balance?.enabled &&
      supportsBalanceCheck[this.options.endpointType ?? this.options.endpoint]
    ) {
      await checkBalance({
@@ -711,32 +672,16 @@ class BaseClient {
    };

    if (typeof completion === 'string') {
-      responseMessage.text = completion;
+      responseMessage.text = addSpaceIfNeeded(generation) + completion;
    } else if (
      Array.isArray(completion) &&
      (this.clientName === EModelEndpoint.agents ||
        isParamEndpoint(this.options.endpoint, this.options.endpointType))
    ) {
      responseMessage.text = '';
-
-      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,
-          );
-        }
-      }
+      responseMessage.content = completion;
    } else if (Array.isArray(completion)) {
-      responseMessage.text = completion.join('');
+      responseMessage.text = addSpaceIfNeeded(generation) + completion.join('');
    }

    if (
@@ -767,14 +712,9 @@ class BaseClient {
      } else {
        responseMessage.tokenCount = this.getTokenCountForResponse(responseMessage);
        completionTokens = responseMessage.tokenCount;
-        await this.recordTokenUsage({
-          usage,
-          promptTokens,
-          completionTokens,
-          balance: balanceConfig,
-          model: responseMessage.model,
-        });
      }
+
+      await this.recordTokenUsage({ promptTokens, completionTokens, usage });
    }

    if (userMessagePromise) {
@@ -852,8 +792,7 @@ class BaseClient {

    userMessage.tokenCount = userMessageTokenCount;
    /*
-      Note: `AgentController` saves the user message if not saved here
-      (noted by `savedMessageIds`), so we update the count of its `userMessage` reference
+      Note: `AskController` saves the user message, so we update the count of its `userMessage` reference
    */
    if (typeof opts?.getReqData === 'function') {
      opts.getReqData({
@@ -862,8 +801,7 @@ class BaseClient {
    }
    /*
      Note: we update the user message to be sure it gets the calculated token count;
-      though `AgentController` saves the user message if not saved here
-      (noted by `savedMessageIds`), EditController does not
+      though `AskController` saves the user message, EditController does not
    */
    await userMessagePromise;
    await this.updateMessageInDatabase({
@@ -1155,50 +1093,6 @@ 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);
@@ -1207,135 +1101,8 @@ class BaseClient {
    return await this.sendCompletion(payload, opts);
  }

-  async addDocuments(message, attachments) {
-    const documentResult = await encodeAndFormatDocuments(
-      this.options.req,
-      attachments,
-      {
-        provider: this.options.agent?.provider,
-        useResponsesApi: this.options.agent?.model_parameters?.useResponsesApi,
-      },
-      getStrategyFunctions,
-    );
-    message.documents =
-      documentResult.documents && documentResult.documents.length
-        ? documentResult.documents
-        : undefined;
-    return documentResult.files;
-  }
-
-  async addVideos(message, attachments) {
-    const videoResult = await encodeAndFormatVideos(
-      this.options.req,
-      attachments,
-      this.options.agent.provider,
-      getStrategyFunctions,
-    );
-    message.videos =
-      videoResult.videos && videoResult.videos.length ? videoResult.videos : undefined;
-    return videoResult.files;
-  }
-
-  async addAudios(message, attachments) {
-    const audioResult = await encodeAndFormatAudios(
-      this.options.req,
-      attachments,
-      this.options.agent.provider,
-      getStrategyFunctions,
-    );
-    message.audios =
-      audioResult.audios && audioResult.audios.length ? audioResult.audios : undefined;
-    return audioResult.files;
-  }
-
-  /**
-   * Extracts text context from attachments and sets it on the message.
-   * This handles text that was already extracted from files (OCR, transcriptions, document text, etc.)
-   * @param {TMessage} message - The message to add context to
-   * @param {MongoFile[]} attachments - Array of file attachments
-   * @returns {Promise<void>}
-   */
-  async addFileContextToMessage(message, attachments) {
-    const fileContext = await extractFileContext({
-      attachments,
-      req: this.options?.req,
-      tokenCountFn: (text) => countTokens(text),
-    });
-
-    if (fileContext) {
-      message.fileContext = fileContext;
-    }
-  }
-
-  async processAttachments(message, attachments) {
-    const categorizedAttachments = {
-      images: [],
-      videos: [],
-      audios: [],
-      documents: [],
-    };
-
-    const allFiles = [];
-
-    for (const file of attachments) {
-      /** @type {FileSources} */
-      const source = file.source ?? FileSources.local;
-      if (source === FileSources.text) {
-        allFiles.push(file);
-        continue;
-      }
-      if (file.embedded === true || file.metadata?.fileIdentifier != null) {
-        allFiles.push(file);
-        continue;
-      }
-
-      if (file.type.startsWith('image/')) {
-        categorizedAttachments.images.push(file);
-      } else if (file.type === 'application/pdf') {
-        categorizedAttachments.documents.push(file);
-        allFiles.push(file);
-      } else if (file.type.startsWith('video/')) {
-        categorizedAttachments.videos.push(file);
-        allFiles.push(file);
-      } else if (file.type.startsWith('audio/')) {
-        categorizedAttachments.audios.push(file);
-        allFiles.push(file);
-      }
-    }
-
-    const [imageFiles] = await Promise.all([
-      categorizedAttachments.images.length > 0
-        ? this.addImageURLs(message, categorizedAttachments.images)
-        : Promise.resolve([]),
-      categorizedAttachments.documents.length > 0
-        ? this.addDocuments(message, categorizedAttachments.documents)
-        : Promise.resolve([]),
-      categorizedAttachments.videos.length > 0
-        ? this.addVideos(message, categorizedAttachments.videos)
-        : Promise.resolve([]),
-      categorizedAttachments.audios.length > 0
-        ? this.addAudios(message, categorizedAttachments.audios)
-        : Promise.resolve([]),
-    ]);
-
-    allFiles.push(...imageFiles);
-
-    const seenFileIds = new Set();
-    const uniqueFiles = [];
-
-    for (const file of allFiles) {
-      if (file.file_id && !seenFileIds.has(file.file_id)) {
-        seenFileIds.add(file.file_id);
-        uniqueFiles.push(file);
-      } else if (!file.file_id) {
-        uniqueFiles.push(file);
-      }
-    }
-
-    return uniqueFiles;
-  }
-
  /**
+   *
   * @param {TMessage[]} _messages
   * @returns {Promise<TMessage[]>}
   */
@@ -1384,8 +1151,7 @@ class BaseClient {
        {},
      );

-      await this.addFileContextToMessage(message, files);
-      await this.processAttachments(message, files);
+      await this.addImageURLs(message, files, this.visionMode);

      this.message_file_map[message.messageId] = files;
      return message;
--- a/api/app/clients/ChatGPTClient.js
+++ b/api/app/clients/ChatGPTClient.js
@@ -0,0 +1,803 @@
+const { Keyv } = require('keyv');
+const crypto = require('crypto');
+const { CohereClient } = require('cohere-ai');
+const { fetchEventSource } = require('@waylaidwanderer/fetch-event-source');
+const { encoding_for_model: encodingForModel, get_encoding: getEncoding } = require('tiktoken');
+const {
+  ImageDetail,
+  EModelEndpoint,
+  resolveHeaders,
+  CohereConstants,
+  mapModelToAzureConfig,
+} = require('librechat-data-provider');
+const { extractBaseURL, constructAzureURL, genAzureChatCompletion } = require('~/utils');
+const { createContextHandlers } = require('./prompts');
+const { createCoherePayload } = require('./llm');
+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;
--- a/api/app/clients/GoogleClient.js
+++ b/api/app/clients/GoogleClient.js
@@ -1,10 +1,6 @@
 const { google } = require('googleapis');
-const { sleep } = require('@librechat/agents');
-const { logger } = require('@librechat/data-schemas');
-const { getModelMaxTokens } = require('@librechat/api');
 const { concat } = require('@langchain/core/utils/stream');
 const { ChatVertexAI } = require('@langchain/google-vertexai');
-const { Tokenizer, getSafetySettings } = require('@librechat/api');
 const { ChatGoogleGenerativeAI } = require('@langchain/google-genai');
 const { GoogleGenerativeAI: GenAI } = require('@google/generative-ai');
 const { HumanMessage, SystemMessage } = require('@langchain/core/messages');
@@ -15,15 +11,19 @@ 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 Tokenizer = require('~/server/services/Tokenizer');
 const { spendTokens } = require('~/models/spendTokens');
+const { getModelMaxTokens } = require('~/utils');
+const { sleep } = require('~/server/utils');
+const { logger } = require('~/config');
 const {
  formatMessage,
  createContextHandlers,
@@ -34,8 +34,7 @@ const BaseClient = require('./BaseClient');

 const loc = process.env.GOOGLE_LOC || 'us-central1';
 const publisher = 'google';
-const endpointPrefix =
-  loc === 'global' ? 'aiplatform.googleapis.com' : `${loc}-aiplatform.googleapis.com`;
+const endpointPrefix = `${loc}-aiplatform.googleapis.com`;

 const settings = endpointSettings[EModelEndpoint.google];
 const EXCLUDED_GENAI_MODELS = /gemini-(?:1\.0|1-0|pro)/;
@@ -166,16 +165,6 @@ 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({
--- a/api/app/clients/OllamaClient.js
+++ b/api/app/clients/OllamaClient.js
@@ -1,11 +1,10 @@
 const { z } = require('zod');
 const axios = require('axios');
 const { Ollama } = require('ollama');
-const { sleep } = require('@librechat/agents');
-const { resolveHeaders } = require('@librechat/api');
-const { logger } = require('@librechat/data-schemas');
 const { Constants } = require('librechat-data-provider');
-const { deriveBaseURL } = require('~/utils');
+const { deriveBaseURL, logAxiosError } = require('~/utils');
+const { sleep } = require('~/server/utils');
+const { logger } = require('~/config');

 const ollamaPayloadSchema = z.object({
  mirostat: z.number().optional(),
@@ -44,7 +43,6 @@ class OllamaClient {
  constructor(options = {}) {
    const host = deriveBaseURL(options.baseURL ?? 'http://localhost:11434');
    this.streamRate = options.streamRate ?? Constants.DEFAULT_STREAM_RATE;
-    this.headers = options.headers ?? {};
    /** @type {Ollama} */
    this.client = new Ollama({ host });
  }
@@ -52,32 +50,27 @@ class OllamaClient {
  /**
   * Fetches Ollama models from the specified base API path.
   * @param {string} baseURL
-   * @param {Object} [options] - Optional configuration
-   * @param {Partial<IUser>} [options.user] - User object for header resolution
-   * @param {Record<string, string>} [options.headers] - Headers to include in the request
   * @returns {Promise<string[]>} The Ollama models.
-   * @throws {Error} Throws if the Ollama API request fails
   */
-  static async fetchModels(baseURL, options = {}) {
+  static async fetchModels(baseURL) {
+    let models = [];
    if (!baseURL) {
+      return models;
+    }
+    try {
+      const ollamaEndpoint = deriveBaseURL(baseURL);
+      /** @type {Promise<AxiosResponse<OllamaListResponse>>} */
+      const response = await axios.get(`${ollamaEndpoint}/api/tags`, {
+        timeout: 5000,
+      });
+      models = response.data.models.map((tag) => tag.name);
+      return models;
+    } catch (error) {
+      const logMessage =
+        "Failed to fetch models from Ollama API. If you are not using Ollama directly, and instead, through some aggregator or reverse proxy that handles fetching via OpenAI spec, ensure the name of the endpoint doesn't start with `ollama` (case-insensitive).";
+      logAxiosError({ message: logMessage, error });
      return [];
    }
-
-    const ollamaEndpoint = deriveBaseURL(baseURL);
-
-    const resolvedHeaders = resolveHeaders({
-      headers: options.headers,
-      user: options.user,
-    });
-
-    /** @type {Promise<AxiosResponse<OllamaListResponse>>} */
-    const response = await axios.get(`${ollamaEndpoint}/api/tags`, {
-      headers: resolvedHeaders,
-      timeout: 5000,
-    });
-
-    const models = response.data.models.map((tag) => tag.name);
-    return models;
  }

  /**
--- a/api/app/clients/OpenAIClient.js
+++ b/api/app/clients/OpenAIClient.js
@@ -1,23 +1,13 @@
-const { logger } = require('@librechat/data-schemas');
+const { OllamaClient } = require('./OllamaClient');
 const { HttpsProxyAgent } = require('https-proxy-agent');
-const { sleep, SplitStreamHandler, CustomOpenAIClient: OpenAI } = require('@librechat/agents');
-const {
-  isEnabled,
-  Tokenizer,
-  createFetch,
-  resolveHeaders,
-  constructAzureURL,
-  getModelMaxTokens,
-  genAzureChatCompletion,
-  getModelMaxOutputTokens,
-  createStreamEventHandlers,
-} = require('@librechat/api');
+const { SplitStreamHandler, CustomOpenAIClient: OpenAI } = require('@librechat/agents');
 const {
  Constants,
  ImageDetail,
  ContentTypes,
  parseTextParts,
  EModelEndpoint,
+  resolveHeaders,
  KnownEndpoints,
  openAISettings,
  ImageDetailCost,
@@ -26,6 +16,13 @@ const {
  validateVisionModel,
  mapModelToAzureConfig,
 } = require('librechat-data-provider');
+const {
+  extractBaseURL,
+  constructAzureURL,
+  getModelMaxTokens,
+  genAzureChatCompletion,
+  getModelMaxOutputTokens,
+} = require('~/utils');
 const {
  truncateText,
  formatMessage,
@@ -34,19 +31,28 @@ const {
  createContextHandlers,
 } = require('./prompts');
 const { encodeAndFormat } = require('~/server/services/Files/images/encode');
+const { createFetch, createStreamEventHandlers } = require('./generators');
+const { addSpaceIfNeeded, isEnabled, sleep } = require('~/server/utils');
+const Tokenizer = require('~/server/services/Tokenizer');
 const { spendTokens } = require('~/models/spendTokens');
-const { addSpaceIfNeeded } = require('~/server/utils');
 const { handleOpenAIErrors } = require('./tools/util');
-const { OllamaClient } = require('./OllamaClient');
+const { createLLM, RunManager } = require('./llm');
+const ChatGPTClient = require('./ChatGPTClient');
 const { summaryBuffer } = require('./memory');
 const { runTitleChain } = require('./chains');
-const { extractBaseURL } = require('~/utils');
 const { tokenSplit } = require('./document');
 const BaseClient = require('./BaseClient');
+const { 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';
@@ -373,12 +379,23 @@ class OpenAIClient extends BaseClient {
    return files;
  }

-  async buildMessages(messages, parentMessageId, { promptPrefix = null }, opts) {
+  async buildMessages(
+    messages,
+    parentMessageId,
+    { isChatCompletion = false, 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;
@@ -613,8 +630,77 @@ class OpenAIClient extends BaseClient {
    return (reply ?? '').trim();
  }

-  initializeLLM() {
-    throw new Error('Deprecated');
+  initializeLLM({
+    model = openAISettings.model.default,
+    modelName,
+    temperature = 0.2,
+    max_tokens,
+    streaming,
+    context,
+    tokenBuffer,
+    initialMessageCount,
+    conversationId,
+  }) {
+    const modelOptions = {
+      modelName: modelName ?? model,
+      temperature,
+      user: this.user,
+    };
+
+    if (max_tokens) {
+      modelOptions.max_tokens = max_tokens;
+    }
+
+    const configOptions = {};
+
+    if (this.langchainProxy) {
+      configOptions.basePath = this.langchainProxy;
+    }
+
+    if (this.useOpenRouter) {
+      configOptions.basePath = 'https://openrouter.ai/api/v1';
+      configOptions.baseOptions = {
+        headers: {
+          'HTTP-Referer': 'https://librechat.ai',
+          'X-Title': 'LibreChat',
+        },
+      };
+    }
+
+    const { headers } = this.options;
+    if (headers && typeof headers === 'object' && !Array.isArray(headers)) {
+      configOptions.baseOptions = {
+        headers: resolveHeaders({
+          ...headers,
+          ...configOptions?.baseOptions?.headers,
+        }),
+      };
+    }
+
+    if (this.options.proxy) {
+      configOptions.httpAgent = new HttpsProxyAgent(this.options.proxy);
+      configOptions.httpsAgent = new HttpsProxyAgent(this.options.proxy);
+    }
+
+    const { req, res, debug } = this.options;
+    const runManager = new RunManager({ req, res, debug, abortController: this.abortController });
+    this.runManager = runManager;
+
+    const llm = createLLM({
+      modelOptions,
+      configOptions,
+      openAIApiKey: this.apiKey,
+      azure: this.azure,
+      streaming,
+      callbacks: runManager.createCallbacks({
+        context,
+        tokenBuffer,
+        conversationId: this.conversationId ?? conversationId,
+        initialMessageCount,
+      }),
+    });
+
+    return llm;
  }

  /**
@@ -632,7 +718,6 @@ class OpenAIClient extends BaseClient {
   *                            In case of failure, it will return the default title, "New Chat".
   */
  async titleConvo({ text, conversationId, responseText = '' }) {
-    const appConfig = this.options.req?.config;
    this.conversationId = conversationId;

    if (this.options.attachments) {
@@ -661,7 +746,8 @@ class OpenAIClient extends BaseClient {
      max_tokens: 16,
    };

-    const azureConfig = appConfig?.endpoints?.[EModelEndpoint.azureOpenAI];
+    /** @type {TAzureConfig | undefined} */
+    const azureConfig = this.options?.req?.app?.locals?.[EModelEndpoint.azureOpenAI];

    const resetTitleOptions = !!(
      (this.azure && azureConfig) ||
@@ -681,7 +767,7 @@ class OpenAIClient extends BaseClient {
        groupMap,
      });

-      this.options.headers = resolveHeaders({ headers });
+      this.options.headers = resolveHeaders(headers);
      this.options.reverseProxyUrl = baseURL ?? null;
      this.langchainProxy = extractBaseURL(this.options.reverseProxyUrl);
      this.apiKey = azureOptions.azureOpenAIApiKey;
@@ -1050,7 +1136,6 @@ ${convo}
  }

  async chatCompletion({ payload, onProgress, abortController = null }) {
-    const appConfig = this.options.req?.config;
    let error = null;
    let intermediateReply = [];
    const errorCallback = (err) => (error = err);
@@ -1074,7 +1159,6 @@ ${convo}
      logger.debug('[OpenAIClient] chatCompletion', { baseURL, modelOptions });
      const opts = {
        baseURL,
-        fetchOptions: {},
      };

      if (this.useOpenRouter) {
@@ -1093,10 +1177,11 @@ ${convo}
      }

      if (this.options.proxy) {
-        opts.fetchOptions.agent = new HttpsProxyAgent(this.options.proxy);
+        opts.httpAgent = new HttpsProxyAgent(this.options.proxy);
      }

-      const azureConfig = appConfig?.endpoints?.[EModelEndpoint.azureOpenAI];
+      /** @type {TAzureConfig | undefined} */
+      const azureConfig = this.options?.req?.app?.locals?.[EModelEndpoint.azureOpenAI];

      if (
        (this.azure && this.isVisionModel && azureConfig) ||
@@ -1113,7 +1198,7 @@ ${convo}
          modelGroupMap,
          groupMap,
        });
-        opts.defaultHeaders = resolveHeaders({ headers });
+        opts.defaultHeaders = resolveHeaders(headers);
        this.langchainProxy = extractBaseURL(baseURL);
        this.apiKey = azureOptions.azureOpenAIApiKey;

@@ -1154,9 +1239,7 @@ ${convo}
      }

      if (this.isOmni === true && modelOptions.max_tokens != null) {
-        const paramName =
-          modelOptions.useResponsesApi === true ? 'max_output_tokens' : 'max_completion_tokens';
-        modelOptions[paramName] = modelOptions.max_tokens;
+        modelOptions.max_completion_tokens = modelOptions.max_tokens;
        delete modelOptions.max_tokens;
      }
      if (this.isOmni === true && modelOptions.temperature != null) {
@@ -1312,7 +1395,7 @@ ${convo}
          ...modelOptions,
          stream: true,
        };
-        const stream = await openai.chat.completions
+        const stream = await openai.beta.chat.completions
          .stream(params)
          .on('abort', () => {
            /* Do nothing here */
--- a/api/app/clients/PluginsClient.js
+++ b/api/app/clients/PluginsClient.js
@@ -0,0 +1,542 @@
+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;
--- a/api/app/clients/TextStream.js
+++ b/api/app/clients/TextStream.js
@@ -1,5 +1,5 @@
 const { Readable } = require('stream');
-const { logger } = require('@librechat/data-schemas');
+const { logger } = require('~/config');

 class TextStream extends Readable {
  constructor(text, options = {}) {
--- a/api/app/clients/agents/CustomAgent/outputParser.js
+++ b/api/app/clients/agents/CustomAgent/outputParser.js
@@ -1,5 +1,5 @@
-const { logger } = require('@librechat/data-schemas');
 const { ZeroShotAgentOutputParser } = require('langchain/agents');
+const { logger } = require('~/config');

 class CustomOutputParser extends ZeroShotAgentOutputParser {
  constructor(fields) {
--- a/api/app/clients/callbacks/createStartHandler.js
+++ b/api/app/clients/callbacks/createStartHandler.js
@@ -0,0 +1,95 @@
+const { promptTokensEstimate } = require('openai-chat-tokens');
+const { EModelEndpoint, supportsBalanceCheck } = require('librechat-data-provider');
+const { formatFromLangChain } = require('~/app/clients/prompts');
+const { getBalanceConfig } = require('~/server/services/Config');
+const { checkBalance } = require('~/models/balanceMethods');
+const { logger } = require('~/config');
+
+const createStartHandler = ({
+  context,
+  conversationId,
+  tokenBuffer = 0,
+  initialMessageCount,
+  manager,
+}) => {
+  return async (_llm, _messages, runId, parentRunId, extraParams) => {
+    const { invocation_params } = extraParams;
+    const { model, functions, function_call } = invocation_params;
+    const messages = _messages[0].map(formatFromLangChain);
+
+    logger.debug(`[createStartHandler] handleChatModelStart: ${context}`, {
+      model,
+      function_call,
+    });
+
+    if (context !== 'title') {
+      logger.debug(`[createStartHandler] handleChatModelStart: ${context}`, {
+        functions,
+      });
+    }
+
+    const payload = { messages };
+    let prelimPromptTokens = 1;
+
+    if (functions) {
+      payload.functions = functions;
+      prelimPromptTokens += 2;
+    }
+
+    if (function_call) {
+      payload.function_call = function_call;
+      prelimPromptTokens -= 5;
+    }
+
+    prelimPromptTokens += promptTokensEstimate(payload);
+    logger.debug('[createStartHandler]', {
+      prelimPromptTokens,
+      tokenBuffer,
+    });
+    prelimPromptTokens += tokenBuffer;
+
+    try {
+      const balance = await getBalanceConfig();
+      if (balance?.enabled && supportsBalanceCheck[EModelEndpoint.openAI]) {
+        const generations =
+          initialMessageCount && messages.length > initialMessageCount
+            ? messages.slice(initialMessageCount)
+            : null;
+        await checkBalance({
+          req: manager.req,
+          res: manager.res,
+          txData: {
+            user: manager.user,
+            tokenType: 'prompt',
+            amount: prelimPromptTokens,
+            debug: manager.debug,
+            generations,
+            model,
+            endpoint: EModelEndpoint.openAI,
+          },
+        });
+      }
+    } catch (err) {
+      logger.error(`[createStartHandler][${context}] checkBalance error`, err);
+      manager.abortController.abort();
+      if (context === 'summary' || context === 'plugins') {
+        manager.addRun(runId, { conversationId, error: err.message });
+        throw new Error(err);
+      }
+      return;
+    }
+
+    manager.addRun(runId, {
+      model,
+      messages,
+      functions,
+      function_call,
+      runId,
+      parentRunId,
+      conversationId,
+      prelimPromptTokens,
+    });
+  };
+};
+
+module.exports = createStartHandler;
--- a/api/app/clients/callbacks/index.js
+++ b/api/app/clients/callbacks/index.js
@@ -0,0 +1,5 @@
+const createStartHandler = require('./createStartHandler');
+
+module.exports = {
+  createStartHandler,
+};
--- a/api/app/clients/chains/runTitleChain.js
+++ b/api/app/clients/chains/runTitleChain.js
@@ -1,7 +1,7 @@
 const { z } = require('zod');
-const { logger } = require('@librechat/data-schemas');
 const { langPrompt, createTitlePrompt, escapeBraces, getSnippet } = require('../prompts');
 const { createStructuredOutputChainFromZod } = require('langchain/chains/openai_functions');
+const { logger } = require('~/config');

 const langSchema = z.object({
  language: z.string().describe('The language of the input text (full noun, no abbreviations).'),
--- a/api/app/clients/generators.js
+++ b/api/app/clients/generators.js
@@ -0,0 +1,71 @@
+const fetch = require('node-fetch');
+const { GraphEvents } = require('@librechat/agents');
+const { logger, sendEvent } = require('~/config');
+const { sleep } = require('~/server/utils');
+
+/**
+ * Makes a function to make HTTP request and logs the process.
+ * @param {Object} params
+ * @param {boolean} [params.directEndpoint] - Whether to use a direct endpoint.
+ * @param {string} [params.reverseProxyUrl] - The reverse proxy URL to use for the request.
+ * @returns {Promise<Response>} - A promise that resolves to the response of the fetch request.
+ */
+function createFetch({ directEndpoint = false, reverseProxyUrl = '' }) {
+  /**
+   * Makes an HTTP request and logs the process.
+   * @param {RequestInfo} url - The URL to make the request to. Can be a string or a Request object.
+   * @param {RequestInit} [init] - Optional init options for the request.
+   * @returns {Promise<Response>} - A promise that resolves to the response of the fetch request.
+   */
+  return async (_url, init) => {
+    let url = _url;
+    if (directEndpoint) {
+      url = reverseProxyUrl;
+    }
+    logger.debug(`Making request to ${url}`);
+    if (typeof Bun !== 'undefined') {
+      return await fetch(url, init);
+    }
+    return await fetch(url, init);
+  };
+}
+
+// Add this at the module level outside the class
+/**
+ * Creates event handlers for stream events that don't capture client references
+ * @param {Object} res - The response object to send events to
+ * @returns {Object} Object containing handler functions
+ */
+function createStreamEventHandlers(res) {
+  return {
+    [GraphEvents.ON_RUN_STEP]: (event) => {
+      if (res) {
+        sendEvent(res, event);
+      }
+    },
+    [GraphEvents.ON_MESSAGE_DELTA]: (event) => {
+      if (res) {
+        sendEvent(res, event);
+      }
+    },
+    [GraphEvents.ON_REASONING_DELTA]: (event) => {
+      if (res) {
+        sendEvent(res, event);
+      }
+    },
+  };
+}
+
+function createHandleLLMNewToken(streamRate) {
+  return async () => {
+    if (streamRate) {
+      await sleep(streamRate);
+    }
+  };
+}
+
+module.exports = {
+  createFetch,
+  createHandleLLMNewToken,
+  createStreamEventHandlers,
+};
--- a/api/app/clients/index.js
+++ b/api/app/clients/index.js
@@ -1,11 +1,15 @@
+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,
--- a/api/app/clients/llm/RunManager.js
+++ b/api/app/clients/llm/RunManager.js
@@ -0,0 +1,105 @@
+const { createStartHandler } = require('~/app/clients/callbacks');
+const { spendTokens } = require('~/models/spendTokens');
+const { logger } = require('~/config');
+
+class RunManager {
+  constructor(fields) {
+    const { req, res, abortController, debug } = fields;
+    this.abortController = abortController;
+    this.user = req.user.id;
+    this.req = req;
+    this.res = res;
+    this.debug = debug;
+    this.runs = new Map();
+    this.convos = new Map();
+  }
+
+  addRun(runId, runData) {
+    if (!this.runs.has(runId)) {
+      this.runs.set(runId, runData);
+      if (runData.conversationId) {
+        this.convos.set(runData.conversationId, runId);
+      }
+      return runData;
+    } else {
+      const existingData = this.runs.get(runId);
+      const update = { ...existingData, ...runData };
+      this.runs.set(runId, update);
+      if (update.conversationId) {
+        this.convos.set(update.conversationId, runId);
+      }
+      return update;
+    }
+  }
+
+  removeRun(runId) {
+    if (this.runs.has(runId)) {
+      this.runs.delete(runId);
+    } else {
+      logger.error(`[api/app/clients/llm/RunManager] Run with ID ${runId} does not exist.`);
+    }
+  }
+
+  getAllRuns() {
+    return Array.from(this.runs.values());
+  }
+
+  getRunById(runId) {
+    return this.runs.get(runId);
+  }
+
+  getRunByConversationId(conversationId) {
+    const runId = this.convos.get(conversationId);
+    return { run: this.runs.get(runId), runId };
+  }
+
+  createCallbacks(metadata) {
+    return [
+      {
+        handleChatModelStart: createStartHandler({ ...metadata, manager: this }),
+        handleLLMEnd: async (output, runId, _parentRunId) => {
+          const { llmOutput, ..._output } = output;
+          logger.debug(`[RunManager] handleLLMEnd: ${JSON.stringify(metadata)}`, {
+            runId,
+            _parentRunId,
+            llmOutput,
+          });
+
+          if (metadata.context !== 'title') {
+            logger.debug('[RunManager] handleLLMEnd:', {
+              output: _output,
+            });
+          }
+
+          const { tokenUsage } = output.llmOutput;
+          const run = this.getRunById(runId);
+          this.removeRun(runId);
+
+          const txData = {
+            user: this.user,
+            model: run?.model ?? 'gpt-3.5-turbo',
+            ...metadata,
+          };
+
+          await spendTokens(txData, tokenUsage);
+        },
+        handleLLMError: async (err) => {
+          logger.error(`[RunManager] handleLLMError: ${JSON.stringify(metadata)}`, err);
+          if (metadata.context === 'title') {
+            return;
+          } else if (metadata.context === 'plugins') {
+            throw new Error(err);
+          }
+          const { conversationId } = metadata;
+          const { run } = this.getRunByConversationId(conversationId);
+          if (run && run.error) {
+            const { error } = run;
+            throw new Error(error);
+          }
+        },
+      },
+    ];
+  }
+}
+
+module.exports = RunManager;
--- a/api/app/clients/llm/createLLM.js
+++ b/api/app/clients/llm/createLLM.js
@@ -0,0 +1,82 @@
+const { ChatOpenAI } = require('@langchain/openai');
+const { sanitizeModelName, constructAzureURL } = require('~/utils');
+const { isEnabled } = require('~/server/utils');
+
+/**
+ * Creates a new instance of a language model (LLM) for chat interactions.
+ *
+ * @param {Object} options - The options for creating the LLM.
+ * @param {ModelOptions} options.modelOptions - The options specific to the model, including modelName, temperature, presence_penalty, frequency_penalty, and other model-related settings.
+ * @param {ConfigOptions} options.configOptions - Configuration options for the API requests, including proxy settings and custom headers.
+ * @param {Callbacks} [options.callbacks] - Callback functions for managing the lifecycle of the LLM, including token buffers, context, and initial message count.
+ * @param {boolean} [options.streaming=false] - Determines if the LLM should operate in streaming mode.
+ * @param {string} options.openAIApiKey - The API key for OpenAI, used for authentication.
+ * @param {AzureOptions} [options.azure={}] - Optional Azure-specific configurations. If provided, Azure configurations take precedence over OpenAI configurations.
+ *
+ * @returns {ChatOpenAI} An instance of the ChatOpenAI class, configured with the provided options.
+ *
+ * @example
+ * const llm = createLLM({
+ *   modelOptions: { modelName: 'gpt-4o-mini', temperature: 0.2 },
+ *   configOptions: { basePath: 'https://example.api/path' },
+ *   callbacks: { onMessage: handleMessage },
+ *   openAIApiKey: 'your-api-key'
+ * });
+ */
+function createLLM({
+  modelOptions,
+  configOptions,
+  callbacks,
+  streaming = false,
+  openAIApiKey,
+  azure = {},
+}) {
+  let credentials = { openAIApiKey };
+  let configuration = {
+    apiKey: openAIApiKey,
+    ...(configOptions.basePath && { baseURL: configOptions.basePath }),
+  };
+
+  /**  @type {AzureOptions} */
+  let azureOptions = {};
+  if (azure) {
+    const useModelName = isEnabled(process.env.AZURE_USE_MODEL_AS_DEPLOYMENT_NAME);
+
+    credentials = {};
+    configuration = {};
+    azureOptions = azure;
+
+    azureOptions.azureOpenAIApiDeploymentName = useModelName
+      ? sanitizeModelName(modelOptions.modelName)
+      : azureOptions.azureOpenAIApiDeploymentName;
+  }
+
+  if (azure && process.env.AZURE_OPENAI_DEFAULT_MODEL) {
+    modelOptions.modelName = process.env.AZURE_OPENAI_DEFAULT_MODEL;
+  }
+
+  if (azure && configOptions.basePath) {
+    const azureURL = constructAzureURL({
+      baseURL: configOptions.basePath,
+      azureOptions,
+    });
+    azureOptions.azureOpenAIBasePath = azureURL.split(
+      `/${azureOptions.azureOpenAIApiDeploymentName}`,
+    )[0];
+  }
+
+  return new ChatOpenAI(
+    {
+      streaming,
+      credentials,
+      configuration,
+      ...azureOptions,
+      ...modelOptions,
+      ...credentials,
+      callbacks,
+    },
+    configOptions,
+  );
+}
+
+module.exports = createLLM;
--- a/api/app/clients/llm/index.js
+++ b/api/app/clients/llm/index.js
@@ -1,5 +1,9 @@
+const createLLM = require('./createLLM');
+const RunManager = require('./RunManager');
 const createCoherePayload = require('./createCoherePayload');

 module.exports = {
+  createLLM,
+  RunManager,
  createCoherePayload,
 };
--- a/api/app/clients/memory/example.js
+++ b/api/app/clients/memory/example.js
@@ -22,17 +22,17 @@
      '\n' +
      'Celestia had the power to grant one wish to anyone who dared to find her abode. Ethan, captivated by the tales he had read and yearning for something greater, approached the cottage with trepidation. When he shared his desire to embark on a grand adventure, Celestia smiled warmly and agreed to grant his wish.\n' +
      '\n' +
-      'With a wave of her wand and a sprinkle of stardust, Celestia bestowed upon Ethan a magical necklace. This necklace, adorned with a rare gemstone called the Eye of Imagination, had the power to turn dreams and imagination into reality. From that moment forward, Ethan\'s every thought and idea became manifest.\n' +
+      "With a wave of her wand and a sprinkle of stardust, Celestia bestowed upon Ethan a magical necklace. This necklace, adorned with a rare gemstone called the Eye of Imagination, had the power to turn dreams and imagination into reality. From that moment forward, Ethan's every thought and idea became manifest.\n" +
      '\n' +
      'Energized by this newfound power, Ethan continued his journey, encountering mythical creatures, solving riddles, and overcoming treacherous obstacles along the way. With the Eye of Imagination, he brought life to ancient statues, unlocked hidden doors, and even tamed fiery dragons.\n' +
      '\n' +
      'As days turned into weeks and weeks into months, Ethan became wiser and more in tune with the world around him. He learned that true adventure was not merely about seeking thrills and conquering the unknown, but also about fostering compassion, friendship, and a deep appreciation for the beauty of the ordinary.\n' +
      '\n' +
-      'Eventually, Ethan\'s journey led him back to his village. With the Eye of Imagination, he transformed the village into a place of wonders and endless possibilities. Fields blossomed into vibrant gardens, simple tools turned into intricate works of art, and the villagers felt a renewed sense of hope and inspiration.\n' +
+      "Eventually, Ethan's journey led him back to his village. With the Eye of Imagination, he transformed the village into a place of wonders and endless possibilities. Fields blossomed into vibrant gardens, simple tools turned into intricate works of art, and the villagers felt a renewed sense of hope and inspiration.\n" +
      '\n' +
-      'Ethan, now known as the Village Magician, realized that the true magic lied within everyone\'s hearts. He taught the villagers to embrace their creativity, to dream big, and to never underestimate the power of imagination. And so, the village flourished, becoming a beacon of wonder and creativity for all to see.\n' +
+      "Ethan, now known as the Village Magician, realized that the true magic lied within everyone's hearts. He taught the villagers to embrace their creativity, to dream big, and to never underestimate the power of imagination. And so, the village flourished, becoming a beacon of wonder and creativity for all to see.\n" +
      '\n' +
-      'In the years that followed, Ethan\'s adventures continued, though mostly within the confines of his beloved village. But he never forgot the thrill of that first grand adventure. And every now and then, when looking up at the starry night sky, he would allow his mind to wander, knowing that the greatest adventures were still waiting to be discovered.',
+      "In the years that followed, Ethan's adventures continued, though mostly within the confines of his beloved village. But he never forgot the thrill of that first grand adventure. And every now and then, when looking up at the starry night sky, he would allow his mind to wander, knowing that the greatest adventures were still waiting to be discovered.",
  },
  {
    role: 'user',
@@ -41,30 +41,30 @@
      '\n' +
      'Once there was a young lad by the name of Ethan, raised in a little hamlet nestled betwixt the verdant knolls, who possessed an irrepressible yearning for knowledge, a thirst unquenchable and a spirit teeming with curiosity. As the golden sun bathed the bucolic land in its effulgent light, he would tread through the village, his ears attuned to the tales spun by the townsfolk, his eyes absorbing the tapestry woven by the world surrounding him.\n' +
      '\n' +
-      'One radiant day, whilst exploring the periphery of the settlement, Ethan chanced upon a timeworn tome, ensconced amidst the roots of an ancient oak, cloaked in the shroud of neglect. The dust gathered upon it spoke of time\'s relentless march. A book of fairy tales – garnished with vivid descriptions of mystical woods, fantastical beasts, and ventures daring beyond the ordinary humdrum existence. Intrigued and beguiled, Ethan pried open the weathered pages and succumbed to their beckoning whispers.\n' +
+      "One radiant day, whilst exploring the periphery of the settlement, Ethan chanced upon a timeworn tome, ensconced amidst the roots of an ancient oak, cloaked in the shroud of neglect. The dust gathered upon it spoke of time's relentless march. A book of fairy tales – garnished with vivid descriptions of mystical woods, fantastical beasts, and ventures daring beyond the ordinary humdrum existence. Intrigued and beguiled, Ethan pried open the weathered pages and succumbed to their beckoning whispers.\n" +
      '\n' +
-      'In each tale, he was transported to a realm of enchantment and wonderment, inexorably tugging at the strings of his yearning for peripatetic exploration. Inspired by the narratives he had devoured, Ethan resolved to bid adieu to kinfolk and embark upon a sojourn, with dreams of procuring a firsthand glimpse into the domain of mystique that lay beyond the village\'s circumscribed boundary.\n' +
+      "In each tale, he was transported to a realm of enchantment and wonderment, inexorably tugging at the strings of his yearning for peripatetic exploration. Inspired by the narratives he had devoured, Ethan resolved to bid adieu to kinfolk and embark upon a sojourn, with dreams of procuring a firsthand glimpse into the domain of mystique that lay beyond the village's circumscribed boundary.\n" +
      '\n' +
      'Thus, he bade tearful farewells, girding himself for a path that guided him to a dense and captivating woodland, whispered of as a sanctuary to mythical beings and clandestine troves of treasures. As Ethan plunged deeper into the heart of the arboreal labyrinth, he felt a palpable surge of electricity, as though the sylvan sentinels whispered enigmatic secrets that only the perceptive ear could discern.\n' +
      '\n' +
-      'It wasn\'t long before his path intertwined with that of a capricious sprite christened Sparkle, bearing an impish grin and eyes sparkling with mischief. Sparkle played the role of Virgil to Ethan\'s Dante, guiding him through the intricate tapestry of arboreal scions, issuing warnings of perils concealed and spinning tales of ancient entities that called this very bosky enclave home.\n' +
+      "It wasn't long before his path intertwined with that of a capricious sprite christened Sparkle, bearing an impish grin and eyes sparkling with mischief. Sparkle played the role of Virgil to Ethan's Dante, guiding him through the intricate tapestry of arboreal scions, issuing warnings of perils concealed and spinning tales of ancient entities that called this very bosky enclave home.\n" +
      '\n' +
      'Together, they stumbled upon a luminous lake, its shimmering waters imbued with a celestial light. At the center lay a diminutive island, upon which reposed a cottage fashioned from tender petals and verdant leaves. It belonged to an ancient sorceress of considerable wisdom, Celestia by name.\n' +
      '\n' +
-      'Celestia, with her power to bestow a single wish on any intrepid soul who happened upon her abode, met Ethan\'s desire with a congenial nod, his fervor for a grand expedition not lost on her penetrating gaze. In response, she bequeathed unto him a necklace of magical manufacture – adorned with the rare gemstone known as the Eye of Imagination – whose very essence transformed dreams into vivid reality. From that moment forward, not a single cogitation nor nebulous fanciful notion of Ethan\'s ever lacked physicality.\n' +
+      "Celestia, with her power to bestow a single wish on any intrepid soul who happened upon her abode, met Ethan's desire with a congenial nod, his fervor for a grand expedition not lost on her penetrating gaze. In response, she bequeathed unto him a necklace of magical manufacture – adorned with the rare gemstone known as the Eye of Imagination – whose very essence transformed dreams into vivid reality. From that moment forward, not a single cogitation nor nebulous fanciful notion of Ethan's ever lacked physicality.\n" +
      '\n' +
      'Energized by this newfound potency, Ethan continued his sojourn, encountering mythical creatures, unraveling cerebral enigmas, and braving perils aplenty along the winding roads of destiny. Armed with the Eye of Imagination, he brought forth life from immobile statuary, unlocked forbidding portals, and even tamed the ferocious beasts of yore – their fiery breath reduced to a whisper.\n' +
      '\n' +
-      'As the weeks metamorphosed into months, Ethan grew wiser and more attuned to the ebb and flow of the world enveloping him. He gleaned that true adventure isn\'t solely confined to sating a thirst for adrenaline and conquering the unknown; indeed, it resides in fostering compassion, fostering amicable bonds, and cherishing the beauty entwined within the quotidian veld.\n' +
+      "As the weeks metamorphosed into months, Ethan grew wiser and more attuned to the ebb and flow of the world enveloping him. He gleaned that true adventure isn't solely confined to sating a thirst for adrenaline and conquering the unknown; indeed, it resides in fostering compassion, fostering amicable bonds, and cherishing the beauty entwined within the quotidian veld.\n" +
      '\n' +
-      'Eventually, Ethan\'s quest drew him homeward, back to his village. Buoying the Eye of Imagination\'s ethereal power, he imbued the hitherto unremarkable settlement with the patina of infinite possibilities. The bounteous fields bloomed into kaleidoscopic gardens, simple instruments transmuting into intricate masterpieces, and the villagers themselves clasped within their hearts a renewed ardor, a conflagration of hope and inspiration.\n' +
+      "Eventually, Ethan's quest drew him homeward, back to his village. Buoying the Eye of Imagination's ethereal power, he imbued the hitherto unremarkable settlement with the patina of infinite possibilities. The bounteous fields bloomed into kaleidoscopic gardens, simple instruments transmuting into intricate masterpieces, and the villagers themselves clasped within their hearts a renewed ardor, a conflagration of hope and inspiration.\n" +
      '\n' +
-      'Behold Ethan, at present hailed as the Village Magician – a cognomen befitting his sorcery wielded within the confines of the community he adored. His exploits may have become tethered to a geographical locale, but the flame kindled by that premier sojourn never flickered. Occasionally, beneath the veiled twilight canopy, when peering skyward toward the astral canvas bedecked with countless luminescent diamonds, he\'d allow his mind to wander, cognizant of boundless adventures yet to take flight, forever awaiting his indomitable spirit.',
+      "Behold Ethan, at present hailed as the Village Magician – a cognomen befitting his sorcery wielded within the confines of the community he adored. His exploits may have become tethered to a geographical locale, but the flame kindled by that premier sojourn never flickered. Occasionally, beneath the veiled twilight canopy, when peering skyward toward the astral canvas bedecked with countless luminescent diamonds, he'd allow his mind to wander, cognizant of boundless adventures yet to take flight, forever awaiting his indomitable spirit.",
  },
  {
    role: 'assistant',
    content:
-      'Beautifully rewritten! It captures the essence of Cormac McCarthy\'s style, with its descriptive language and evocative imagery. The story retains the spirit of adventure and the power of imagination seen in the original. Well done! If you have any more requests or if there\'s anything else I can assist you with, feel free to let me know.',
+      "Beautifully rewritten! It captures the essence of Cormac McCarthy's style, with its descriptive language and evocative imagery. The story retains the spirit of adventure and the power of imagination seen in the original. Well done! If you have any more requests or if there's anything else I can assist you with, feel free to let me know.",
  },
  {
    role: 'user',
@@ -73,19 +73,19 @@
  {
    role: 'assistant',
    content:
-      'Certainly! Here\'s a short story reflecting Cormac McCarthy\'s style:\n' +
+      "Certainly! Here's a short story reflecting Cormac McCarthy's style:\n" +
      '\n' +
      'In a desolate wasteland scarred by the sins of mankind, a lone traveler ventured forth, his name forgotten by time, etched upon the shadows of memory. He navigated the barren plains, where the sun beat down, merciless and unrelenting, scorching the parched earth beneath his weary boots.\n' +
      '\n' +
      'Desolation was his only companion, the forsaken landscape a testament to the darkness that had befallen this forsaken realm. The wind whispered secrets long lost, carrying with it the echoes of a civilization long since undone, swallowed by the abyss.\n' +
      '\n' +
-      'Amongst the shattered remnants of fallen cities, he sought solace, a glimmer of hope amidst the desolation. Rusted remnants of steel and concrete stood as sentinels of mankind\'s hubris, their jagged edges a chilling reminder of the fragility of all that was once deemed invincible.\n' +
+      "Amongst the shattered remnants of fallen cities, he sought solace, a glimmer of hope amidst the desolation. Rusted remnants of steel and concrete stood as sentinels of mankind's hubris, their jagged edges a chilling reminder of the fragility of all that was once deemed invincible.\n" +
      '\n' +
      'As twilight descended, casting a pallid hue upon the lifeless land, he happened upon a dilapidated house, its walls crumbling under the weight of time. Drawn to the desolation, he pushed open the creaking door, its rusty hinges screaming in protest.\n' +
      '\n' +
      'Inside, a world frozen in time awaited him. Dust particles danced through the air, illuminated by slivers of light that pierced the broken windows. Here, the past lay trapped, a melancholic tableau of lives lost and dreams shattered.\n' +
      '\n' +
-      'He wandered through the empty rooms, bare walls whispering stories untold. In the corner, a dilapidated piano stood, its ivory keys long since abandoned by a musician\'s touch. He approached it cautiously, a faint glimmer of anticipation igniting in his weather-worn eyes.\n' +
+      "He wandered through the empty rooms, bare walls whispering stories untold. In the corner, a dilapidated piano stood, its ivory keys long since abandoned by a musician's touch. He approached it cautiously, a faint glimmer of anticipation igniting in his weather-worn eyes.\n" +
      '\n' +
      'Determined, he carefully brushed away the dust and debris that cloaked the instrument. With trembling hands, he pressed a key, and a discordant note filled the air, mingling with the sighs of forgotten melodies. The sound echoed through the silence, bringing with it the remnants of forgotten symphonies.\n' +
      '\n' +
@@ -103,13 +103,13 @@
  {
    role: 'assistant',
    content:
-      'Thank you! I\'m glad you enjoyed it. If you have any more writing requests or if there\'s anything else I can assist you with, feel free to let me know.',
+      "Thank you! I'm glad you enjoyed it. If you have any more writing requests or if there's anything else I can assist you with, feel free to let me know.",
  },
  { role: 'user', content: 'you are very helpful' },
  {
    role: 'assistant',
    content:
-      'Thank you for your kind words! I\'m here to assist you in any way I can. If you have any more questions, need further assistance, or just want to chat, feel free to reach out.',
+      "Thank you for your kind words! I'm here to assist you in any way I can. If you have any more questions, need further assistance, or just want to chat, feel free to reach out.",
  },
  { role: 'user', content: 'no you man' },
 ];
--- a/api/app/clients/memory/summaryBuffer.demo.js
+++ b/api/app/clients/memory/summaryBuffer.demo.js
@@ -0,0 +1,31 @@
+require('dotenv').config();
+const { ChatOpenAI } = require('@langchain/openai');
+const { getBufferString, ConversationSummaryBufferMemory } = require('langchain/memory');
+
+const chatPromptMemory = new ConversationSummaryBufferMemory({
+  llm: new ChatOpenAI({ modelName: 'gpt-4o-mini', temperature: 0 }),
+  maxTokenLimit: 10,
+  returnMessages: true,
+});
+
+(async () => {
+  await chatPromptMemory.saveContext({ input: "hi my name's Danny" }, { output: 'whats up' });
+  await chatPromptMemory.saveContext({ input: 'not much you' }, { output: 'not much' });
+  await chatPromptMemory.saveContext(
+    { input: 'are you excited for the olympics?' },
+    { output: 'not really' },
+  );
+
+  // We can also utilize the predict_new_summary method directly.
+  const messages = await chatPromptMemory.chatHistory.getMessages();
+  console.log('MESSAGES\n\n');
+  console.log(JSON.stringify(messages));
+  const previous_summary = '';
+  const predictSummary = await chatPromptMemory.predictNewSummary(messages, previous_summary);
+  console.log('SUMMARY\n\n');
+  console.log(JSON.stringify(getBufferString([{ role: 'system', content: predictSummary }])));
+
+  // const { history } = await chatPromptMemory.loadMemoryVariables({});
+  // console.log('HISTORY\n\n');
+  // console.log(JSON.stringify(history));
+})();
--- a/api/app/clients/memory/summaryBuffer.js
+++ b/api/app/clients/memory/summaryBuffer.js
@@ -1,7 +1,7 @@
-const { logger } = require('@librechat/data-schemas');
 const { ConversationSummaryBufferMemory, ChatMessageHistory } = require('langchain/memory');
 const { formatLangChainMessages, SUMMARY_PROMPT } = require('../prompts');
 const { predictNewSummary } = require('../chains');
+const { logger } = require('~/config');

 const createSummaryBufferMemory = ({ llm, prompt, messages, ...rest }) => {
  const chatHistory = new ChatMessageHistory(messages);
--- a/api/app/clients/output_parsers/addImages.js
+++ b/api/app/clients/output_parsers/addImages.js
@@ -1,4 +1,4 @@
-const { logger } = require('@librechat/data-schemas');
+const { logger } = require('~/config');

 /**
 * The `addImages` function corrects any erroneous image URLs in the `responseMessage.text`
--- a/api/app/clients/output_parsers/addImages.spec.js
+++ b/api/app/clients/output_parsers/addImages.spec.js
@@ -74,7 +74,7 @@ describe('addImages', () => {

  it('should append correctly from a real scenario', () => {
    responseMessage.text =
-      'Here is the generated image based on your request. It depicts a surreal landscape filled with floating musical notes. The style is impressionistic, with vibrant sunset hues dominating the scene. At the center, there\'s a silhouette of a grand piano, adding a dreamy emotion to the overall image. This could serve as a unique and creative music album cover. Would you like to make any changes or generate another image?';
+      "Here is the generated image based on your request. It depicts a surreal landscape filled with floating musical notes. The style is impressionistic, with vibrant sunset hues dominating the scene. At the center, there's a silhouette of a grand piano, adding a dreamy emotion to the overall image. This could serve as a unique and creative music album cover. Would you like to make any changes or generate another image?";
    const originalText = responseMessage.text;
    const imageMarkdown = '![generated image](/images/img-RnVWaYo2Yg4x3e0isICiMuf5.png)';
    intermediateSteps.push({ observation: imageMarkdown });
--- a/api/app/clients/output_parsers/handleOutputs.js
+++ b/api/app/clients/output_parsers/handleOutputs.js
@@ -65,14 +65,14 @@ function buildPromptPrefix({ result, message, functionsAgent }) {
  const preliminaryAnswer =
    result.output?.length > 0 ? `Preliminary Answer: "${result.output.trim()}"` : '';
  const prefix = preliminaryAnswer
-    ? 'review and improve the answer you generated using plugins in response to the User Message below. The user hasn\'t seen your answer or thoughts yet.'
+    ? "review and improve the answer you generated using plugins in response to the User Message below. The user hasn't seen your answer or thoughts yet."
    : 'respond to the User Message below based on your preliminary thoughts & actions.';

  return `As a helpful AI Assistant, ${prefix}${errorMessage}\n${internalActions}
 ${preliminaryAnswer}
 Reply conversationally to the User based on your ${
-  preliminaryAnswer ? 'preliminary answer, ' : ''
-}internal actions, thoughts, and observations, making improvements wherever possible, but do not modify URLs.
+    preliminaryAnswer ? 'preliminary answer, ' : ''
+  }internal actions, thoughts, and observations, making improvements wherever possible, but do not modify URLs.
 ${
  preliminaryAnswer
    ? ''
--- a/api/app/clients/prompts/addCacheControl.spec.js
+++ b/api/app/clients/prompts/addCacheControl.spec.js
@@ -6,7 +6,7 @@ describe('addCacheControl', () => {
      { role: 'user', content: [{ type: 'text', text: 'Hello' }] },
      { role: 'assistant', content: [{ type: 'text', text: 'Hi there' }] },
      { role: 'user', content: [{ type: 'text', text: 'How are you?' }] },
-      { role: 'assistant', content: [{ type: 'text', text: 'I\'m doing well, thanks!' }] },
+      { role: 'assistant', content: [{ type: 'text', text: "I'm doing well, thanks!" }] },
      { role: 'user', content: [{ type: 'text', text: 'Great!' }] },
    ];

@@ -22,7 +22,7 @@ describe('addCacheControl', () => {
      { role: 'user', content: 'Hello' },
      { role: 'assistant', content: 'Hi there' },
      { role: 'user', content: 'How are you?' },
-      { role: 'assistant', content: 'I\'m doing well, thanks!' },
+      { role: 'assistant', content: "I'm doing well, thanks!" },
      { role: 'user', content: 'Great!' },
    ];

@@ -140,7 +140,7 @@ describe('addCacheControl', () => {
      { role: 'user', content: 'Hello' },
      { role: 'assistant', content: 'Hi there' },
      { role: 'user', content: [{ type: 'text', text: 'How are you?' }] },
-      { role: 'assistant', content: 'I\'m doing well, thanks!' },
+      { role: 'assistant', content: "I'm doing well, thanks!" },
      { role: 'user', content: 'Great!' },
    ];

@@ -160,7 +160,7 @@ describe('addCacheControl', () => {
      },
    ]);
    expect(result[1].content).toBe('Hi there');
-    expect(result[3].content).toBe('I\'m doing well, thanks!');
+    expect(result[3].content).toBe("I'm doing well, thanks!");
  });

  test('should handle edge case with multiple content types', () => {
--- a/api/app/clients/prompts/artifacts.js
+++ b/api/app/clients/prompts/artifacts.js
@@ -3,8 +3,6 @@ const { EModelEndpoint, ArtifactModes } = require('librechat-data-provider');
 const { generateShadcnPrompt } = require('~/app/clients/prompts/shadcn-docs/generate');
 const { components } = require('~/app/clients/prompts/shadcn-docs/components');

-/** @deprecated */
-// eslint-disable-next-line no-unused-vars
 const artifactsPromptV1 = dedent`The assistant can create and reference artifacts during conversations.
  
 Artifacts are for substantial, self-contained content that users might modify or reuse, displayed in a separate UI window for clarity.
@@ -116,7 +114,6 @@ Here are some examples of correct usage of artifacts:
    </assistant_response>
  </example>
 </examples>`;
-
 const artifactsPrompt = dedent`The assistant can create and reference artifacts during conversations.
  
 Artifacts are for substantial, self-contained content that users might modify or reuse, displayed in a separate UI window for clarity.
@@ -167,10 +164,6 @@ Artifacts are for substantial, self-contained content that users might modify or
    - SVG: "image/svg+xml"
      - The user interface will render the Scalable Vector Graphics (SVG) image within the artifact tags.
      - The assistant should specify the viewbox of the SVG rather than defining a width/height
-    - Markdown: "text/markdown" or "text/md"
-      - The user interface will render Markdown content placed within the artifact tags.
-      - Supports standard Markdown syntax including headers, lists, links, images, code blocks, tables, and more.
-      - Both "text/markdown" and "text/md" are accepted as valid MIME types for Markdown content.
    - Mermaid Diagrams: "application/vnd.mermaid"
      - The user interface will render Mermaid diagrams placed within the artifact tags.
    - React Components: "application/vnd.react"
@@ -372,10 +365,6 @@ Artifacts are for substantial, self-contained content that users might modify or
    - SVG: "image/svg+xml"
      - The user interface will render the Scalable Vector Graphics (SVG) image within the artifact tags.
      - The assistant should specify the viewbox of the SVG rather than defining a width/height
-    - Markdown: "text/markdown" or "text/md"
-      - The user interface will render Markdown content placed within the artifact tags.
-      - Supports standard Markdown syntax including headers, lists, links, images, code blocks, tables, and more.
-      - Both "text/markdown" and "text/md" are accepted as valid MIME types for Markdown content.
    - Mermaid Diagrams: "application/vnd.mermaid"
      - The user interface will render Mermaid diagrams placed within the artifact tags.
    - React Components: "application/vnd.react"
--- a/api/app/clients/prompts/createContextHandlers.js
+++ b/api/app/clients/prompts/createContextHandlers.js
@@ -1,6 +1,6 @@
 const axios = require('axios');
-const { logger } = require('@librechat/data-schemas');
-const { isEnabled, generateShortLivedToken } = require('@librechat/api');
+const { isEnabled } = require('~/server/utils');
+const { logger } = require('~/config');

 const footer = `Use the context as your learned knowledge to better answer the user.

@@ -18,7 +18,7 @@ function createContextHandlers(req, userMessageContent) {
  const queryPromises = [];
  const processedFiles = [];
  const processedIds = new Set();
-  const jwtToken = generateShortLivedToken(req.user.id);
+  const jwtToken = req.headers.authorization.split(' ')[1];
  const useFullContext = isEnabled(process.env.RAG_USE_FULL_CONTEXT);

  const query = async (file) => {
--- a/api/app/clients/prompts/formatAgentMessages.spec.js
+++ b/api/app/clients/prompts/formatAgentMessages.spec.js
@@ -130,7 +130,7 @@ describe('formatAgentMessages', () => {
        content: [
          {
            type: ContentTypes.TEXT,
-            [ContentTypes.TEXT]: 'I\'ll search for that information.',
+            [ContentTypes.TEXT]: "I'll search for that information.",
            tool_call_ids: ['search_1'],
          },
          {
@@ -144,7 +144,7 @@ describe('formatAgentMessages', () => {
          },
          {
            type: ContentTypes.TEXT,
-            [ContentTypes.TEXT]: 'Now, I\'ll convert the temperature.',
+            [ContentTypes.TEXT]: "Now, I'll convert the temperature.",
            tool_call_ids: ['convert_1'],
          },
          {
@@ -156,7 +156,7 @@ describe('formatAgentMessages', () => {
              output: '23.89°C',
            },
          },
-          { type: ContentTypes.TEXT, [ContentTypes.TEXT]: 'Here\'s your answer.' },
+          { type: ContentTypes.TEXT, [ContentTypes.TEXT]: "Here's your answer." },
        ],
      },
    ];
@@ -171,7 +171,7 @@ describe('formatAgentMessages', () => {
    expect(result[4]).toBeInstanceOf(AIMessage);

    // Check first AIMessage
-    expect(result[0].content).toBe('I\'ll search for that information.');
+    expect(result[0].content).toBe("I'll search for that information.");
    expect(result[0].tool_calls).toHaveLength(1);
    expect(result[0].tool_calls[0]).toEqual({
      id: 'search_1',
@@ -187,7 +187,7 @@ describe('formatAgentMessages', () => {
    );

    // Check second AIMessage
-    expect(result[2].content).toBe('Now, I\'ll convert the temperature.');
+    expect(result[2].content).toBe("Now, I'll convert the temperature.");
    expect(result[2].tool_calls).toHaveLength(1);
    expect(result[2].tool_calls[0]).toEqual({
      id: 'convert_1',
@@ -202,7 +202,7 @@ describe('formatAgentMessages', () => {

    // Check final AIMessage
    expect(result[4].content).toStrictEqual([
-      { [ContentTypes.TEXT]: 'Here\'s your answer.', type: ContentTypes.TEXT },
+      { [ContentTypes.TEXT]: "Here's your answer.", type: ContentTypes.TEXT },
    ]);
  });

@@ -217,7 +217,7 @@ describe('formatAgentMessages', () => {
        role: 'assistant',
        content: [{ type: ContentTypes.TEXT, [ContentTypes.TEXT]: 'How can I help you?' }],
      },
-      { role: 'user', content: 'What\'s the weather?' },
+      { role: 'user', content: "What's the weather?" },
      {
        role: 'assistant',
        content: [
@@ -240,7 +240,7 @@ describe('formatAgentMessages', () => {
      {
        role: 'assistant',
        content: [
-          { type: ContentTypes.TEXT, [ContentTypes.TEXT]: 'Here\'s the weather information.' },
+          { type: ContentTypes.TEXT, [ContentTypes.TEXT]: "Here's the weather information." },
        ],
      },
    ];
@@ -265,12 +265,12 @@ describe('formatAgentMessages', () => {
      { [ContentTypes.TEXT]: 'How can I help you?', type: ContentTypes.TEXT },
    ]);
    expect(result[2].content).toStrictEqual([
-      { [ContentTypes.TEXT]: 'What\'s the weather?', type: ContentTypes.TEXT },
+      { [ContentTypes.TEXT]: "What's the weather?", type: ContentTypes.TEXT },
    ]);
    expect(result[3].content).toBe('Let me check that for you.');
    expect(result[4].content).toBe('Sunny, 75°F');
    expect(result[5].content).toStrictEqual([
-      { [ContentTypes.TEXT]: 'Here\'s the weather information.', type: ContentTypes.TEXT },
+      { [ContentTypes.TEXT]: "Here's the weather information.", type: ContentTypes.TEXT },
    ]);

    // Check that there are no consecutive AIMessages
--- a/api/app/clients/prompts/formatMessages.js
+++ b/api/app/clients/prompts/formatMessages.js
@@ -237,9 +237,41 @@ 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,
 };
--- a/api/app/clients/prompts/instructions.js
+++ b/api/app/clients/prompts/instructions.js
@@ -1,8 +1,8 @@
 module.exports = {
  instructions:
-    'Remember, all your responses MUST be in the format described. Do not respond unless it\'s in the format described, using the structure of Action, Action Input, etc.',
+    "Remember, all your responses MUST be in the format described. Do not respond unless it's in the format described, using the structure of Action, Action Input, etc.",
  errorInstructions:
-    '\nYou encountered an error in attempting a response. The user is not aware of the error so you shouldn\'t mention it.\nReview the actions taken carefully in case there is a partial or complete answer within them.\nError Message:',
+    "\nYou encountered an error in attempting a response. The user is not aware of the error so you shouldn't mention it.\nReview the actions taken carefully in case there is a partial or complete answer within them.\nError Message:",
  imageInstructions:
    'You must include the exact image paths from above, formatted in Markdown syntax: ![alt-text](URL)',
  completionInstructions:
--- a/api/app/clients/prompts/shadcn-docs/generate.js
+++ b/api/app/clients/prompts/shadcn-docs/generate.js
@@ -18,17 +18,17 @@ function generateShadcnPrompt(options) {
    Here are the components that are available, along with how to import them, and how to use them:

    ${Object.values(components)
-    .map((component) => {
-      if (useXML) {
-        return dedent`
+      .map((component) => {
+        if (useXML) {
+          return dedent`
            <component>
              <name>${component.componentName}</name>
              <import-instructions>${component.importDocs}</import-instructions>
              <usage-instructions>${component.usageDocs}</usage-instructions>
            </component>
          `;
-      } else {
-        return dedent`
+        } else {
+          return dedent`
            # ${component.componentName}

            ## Import Instructions
@@ -37,9 +37,9 @@ function generateShadcnPrompt(options) {
            ## Usage Instructions
            ${component.usageDocs}
          `;
-      }
-    })
-    .join('\n\n')}
+        }
+      })
+      .join('\n\n')}
  `;

  return systemPrompt;
--- a/api/app/clients/specs/AnthropicClient.test.js
+++ b/api/app/clients/specs/AnthropicClient.test.js
@@ -245,7 +245,7 @@ describe('AnthropicClient', () => {
    });

    describe('Claude 4 model headers', () => {
-      it('should add "prompt-caching" and "context-1m" beta headers for claude-sonnet-4 model', () => {
+      it('should add "prompt-caching" beta header for claude-sonnet-4 model', () => {
        const client = new AnthropicClient('test-api-key');
        const modelOptions = {
          model: 'claude-sonnet-4-20250514',
@@ -255,30 +255,10 @@ describe('AnthropicClient', () => {
        expect(anthropicClient._options.defaultHeaders).toBeDefined();
        expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
        expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
-          'prompt-caching-2024-07-31,context-1m-2025-08-07',
+          'prompt-caching-2024-07-31',
        );
      });

-      it('should add "prompt-caching" and "context-1m" beta headers for claude-sonnet-4 model formats', () => {
-        const client = new AnthropicClient('test-api-key');
-        const modelVariations = [
-          'claude-sonnet-4-20250514',
-          'claude-sonnet-4-latest',
-          'anthropic/claude-sonnet-4-20250514',
-        ];
-
-        modelVariations.forEach((model) => {
-          const modelOptions = { model };
-          client.setOptions({ modelOptions, promptCache: true });
-          const anthropicClient = client.getClient(modelOptions);
-          expect(anthropicClient._options.defaultHeaders).toBeDefined();
-          expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
-          expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
-            'prompt-caching-2024-07-31,context-1m-2025-08-07',
-          );
-        });
-      });
-
      it('should add "prompt-caching" beta header for claude-opus-4 model', () => {
        const client = new AnthropicClient('test-api-key');
        const modelOptions = {
@@ -293,6 +273,20 @@ describe('AnthropicClient', () => {
        );
      });

+      it('should add "prompt-caching" beta header for claude-4-sonnet model', () => {
+        const client = new AnthropicClient('test-api-key');
+        const modelOptions = {
+          model: 'claude-4-sonnet-20250514',
+        };
+        client.setOptions({ modelOptions, promptCache: true });
+        const anthropicClient = client.getClient(modelOptions);
+        expect(anthropicClient._options.defaultHeaders).toBeDefined();
+        expect(anthropicClient._options.defaultHeaders).toHaveProperty('anthropic-beta');
+        expect(anthropicClient._options.defaultHeaders['anthropic-beta']).toBe(
+          'prompt-caching-2024-07-31',
+        );
+      });
+
      it('should add "prompt-caching" beta header for claude-4-opus model', () => {
        const client = new AnthropicClient('test-api-key');
        const modelOptions = {
@@ -315,7 +309,7 @@ describe('AnthropicClient', () => {
      };
      client.setOptions({ modelOptions, promptCache: true });
      const anthropicClient = client.getClient(modelOptions);
-      expect(anthropicClient._options.defaultHeaders).toBeUndefined();
+      expect(anthropicClient.defaultHeaders).not.toHaveProperty('anthropic-beta');
    });

    it('should not add beta header for other models', () => {
@@ -326,7 +320,7 @@ describe('AnthropicClient', () => {
        },
      });
      const anthropicClient = client.getClient();
-      expect(anthropicClient._options.defaultHeaders).toBeUndefined();
+      expect(anthropicClient.defaultHeaders).not.toHaveProperty('anthropic-beta');
    });
  });

--- a/api/app/clients/specs/BaseClient.test.js
+++ b/api/app/clients/specs/BaseClient.test.js
@@ -1,15 +1,7 @@
 const { Constants } = require('librechat-data-provider');
 const { initializeFakeClient } = require('./FakeClient');

-jest.mock('~/db/connect');
-jest.mock('~/server/services/Config', () => ({
-  getAppConfig: jest.fn().mockResolvedValue({
-    // Default app config for tests
-    paths: { uploads: '/tmp' },
-    fileStrategy: 'local',
-    memory: { disabled: false },
-  }),
-}));
+jest.mock('~/lib/db/connectDb');
 jest.mock('~/models', () => ({
  User: jest.fn(),
  Key: jest.fn(),
@@ -41,9 +33,7 @@ jest.mock('~/models', () => ({
 const { getConvo, saveConvo } = require('~/models');

 jest.mock('@librechat/agents', () => {
-  const { Providers } = jest.requireActual('@librechat/agents');
  return {
-    Providers,
    ChatOpenAI: jest.fn().mockImplementation(() => {
      return {};
    }),
@@ -430,46 +420,6 @@ 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 = {
@@ -587,8 +537,6 @@ describe('BaseClient', () => {
      expect(onStart).toHaveBeenCalledWith(
        expect.objectContaining({ text: 'Hello, world!' }),
        expect.any(String),
-        /** `isNewConvo` */
-        true,
      );
    });

--- a/api/app/clients/specs/FakeClient.js
+++ b/api/app/clients/specs/FakeClient.js
@@ -1,5 +1,5 @@
-const { getModelMaxTokens } = require('@librechat/api');
 const BaseClient = require('../BaseClient');
+const { getModelMaxTokens } = require('../../../utils');

 class FakeClient extends BaseClient {
  constructor(apiKey, options = {}) {
--- a/api/app/clients/specs/OpenAIClient.test.js
+++ b/api/app/clients/specs/OpenAIClient.test.js
@@ -5,7 +5,7 @@ const getLogStores = require('~/cache/getLogStores');
 const OpenAIClient = require('../OpenAIClient');
 jest.mock('meilisearch');

-jest.mock('~/db/connect');
+jest.mock('~/lib/db/connectDb');
 jest.mock('~/models', () => ({
  User: jest.fn(),
  Key: jest.fn(),
@@ -531,6 +531,44 @@ 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' }];
--- a/api/app/clients/specs/PluginsClient.test.js
+++ b/api/app/clients/specs/PluginsClient.test.js
@@ -0,0 +1,314 @@
+const crypto = require('crypto');
+const { Constants } = require('librechat-data-provider');
+const { HumanMessage, AIMessage } = require('@langchain/core/messages');
+const PluginsClient = require('../PluginsClient');
+
+jest.mock('~/lib/db/connectDb');
+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));
+    });
+  });
+});
--- a/api/app/clients/tools/dynamic/OpenAPIPlugin.js
+++ b/api/app/clients/tools/dynamic/OpenAPIPlugin.js
@@ -0,0 +1,184 @@
+require('dotenv').config();
+const fs = require('fs');
+const { z } = require('zod');
+const path = require('path');
+const yaml = require('js-yaml');
+const { createOpenAPIChain } = require('langchain/chains');
+const { DynamicStructuredTool } = require('@langchain/core/tools');
+const { ChatPromptTemplate, HumanMessagePromptTemplate } = require('@langchain/core/prompts');
+const { logger } = require('~/config');
+
+function addLinePrefix(text, prefix = '// ') {
+  return text
+    .split('\n')
+    .map((line) => prefix + line)
+    .join('\n');
+}
+
+function createPrompt(name, functions) {
+  const prefix = `// The ${name} tool has the following functions. Determine the desired or most optimal function for the user's query:`;
+  const functionDescriptions = functions
+    .map((func) => `// - ${func.name}: ${func.description}`)
+    .join('\n');
+  return `${prefix}\n${functionDescriptions}
+// You are an expert manager and scrum master. You must provide a detailed intent to better execute the function.
+// Always format as such: {{"func": "function_name", "intent": "intent and expected result"}}`;
+}
+
+const AuthBearer = z
+  .object({
+    type: z.string().includes('service_http'),
+    authorization_type: z.string().includes('bearer'),
+    verification_tokens: z.object({
+      openai: z.string(),
+    }),
+  })
+  .catch(() => false);
+
+const AuthDefinition = z
+  .object({
+    type: z.string(),
+    authorization_type: z.string(),
+    verification_tokens: z.object({
+      openai: z.string(),
+    }),
+  })
+  .catch(() => false);
+
+async function readSpecFile(filePath) {
+  try {
+    const fileContents = await fs.promises.readFile(filePath, 'utf8');
+    if (path.extname(filePath) === '.json') {
+      return JSON.parse(fileContents);
+    }
+    return yaml.load(fileContents);
+  } catch (e) {
+    logger.error('[readSpecFile] error', e);
+    return false;
+  }
+}
+
+async function getSpec(url) {
+  const RegularUrl = z
+    .string()
+    .url()
+    .catch(() => false);
+
+  if (RegularUrl.parse(url) && path.extname(url) === '.json') {
+    const response = await fetch(url);
+    return await response.json();
+  }
+
+  const ValidSpecPath = z
+    .string()
+    .url()
+    .catch(async () => {
+      const spec = path.join(__dirname, '..', '.well-known', 'openapi', url);
+      if (!fs.existsSync(spec)) {
+        return false;
+      }
+
+      return await readSpecFile(spec);
+    });
+
+  return ValidSpecPath.parse(url);
+}
+
+async function createOpenAPIPlugin({ data, llm, user, message, memory, signal }) {
+  let spec;
+  try {
+    spec = await getSpec(data.api.url);
+  } catch (error) {
+    logger.error('[createOpenAPIPlugin] getSpec error', error);
+    return null;
+  }
+
+  if (!spec) {
+    logger.warn('[createOpenAPIPlugin] No spec found');
+    return null;
+  }
+
+  const headers = {};
+  const { auth, name_for_model, description_for_model, description_for_human } = data;
+  if (auth && AuthDefinition.parse(auth)) {
+    logger.debug('[createOpenAPIPlugin] auth detected', auth);
+    const { openai } = auth.verification_tokens;
+    if (AuthBearer.parse(auth)) {
+      headers.authorization = `Bearer ${openai}`;
+      logger.debug('[createOpenAPIPlugin] added auth bearer', headers);
+    }
+  }
+
+  const chainOptions = { llm };
+
+  if (data.headers && data.headers['librechat_user_id']) {
+    logger.debug('[createOpenAPIPlugin] id detected', headers);
+    headers[data.headers['librechat_user_id']] = user;
+  }
+
+  if (Object.keys(headers).length > 0) {
+    logger.debug('[createOpenAPIPlugin] headers detected', headers);
+    chainOptions.headers = headers;
+  }
+
+  if (data.params) {
+    logger.debug('[createOpenAPIPlugin] params detected', data.params);
+    chainOptions.params = data.params;
+  }
+
+  let history = '';
+  if (memory) {
+    logger.debug('[createOpenAPIPlugin] openAPI chain: memory detected', memory);
+    const { history: chat_history } = await memory.loadMemoryVariables({});
+    history = chat_history?.length > 0 ? `\n\n## Chat History:\n${chat_history}\n` : '';
+  }
+
+  chainOptions.prompt = ChatPromptTemplate.fromMessages([
+    HumanMessagePromptTemplate.fromTemplate(
+      `# Use the provided API's to respond to this query:\n\n{query}\n\n## Instructions:\n${addLinePrefix(
+        description_for_model,
+      )}${history}`,
+    ),
+  ]);
+
+  const chain = await createOpenAPIChain(spec, chainOptions);
+
+  const { functions } = chain.chains[0].lc_kwargs.llmKwargs;
+
+  return new DynamicStructuredTool({
+    name: name_for_model,
+    description_for_model: `${addLinePrefix(description_for_human)}${createPrompt(
+      name_for_model,
+      functions,
+    )}`,
+    description: `${description_for_human}`,
+    schema: z.object({
+      func: z
+        .string()
+        .describe(
+          `The function to invoke. The functions available are: ${functions
+            .map((func) => func.name)
+            .join(', ')}`,
+        ),
+      intent: z
+        .string()
+        .describe('Describe your intent with the function and your expected result'),
+    }),
+    func: async ({ func = '', intent = '' }) => {
+      const filteredFunctions = functions.filter((f) => f.name === func);
+      chain.chains[0].lc_kwargs.llmKwargs.functions = filteredFunctions;
+      const query = `${message}${func?.length > 0 ? `\n// Intent: ${intent}` : ''}`;
+      const result = await chain.call({
+        query,
+        signal,
+      });
+      return result.response;
+    },
+  });
+}
+
+module.exports = {
+  getSpec,
+  readSpecFile,
+  createOpenAPIPlugin,
+};
--- a/api/app/clients/tools/dynamic/OpenAPIPlugin.spec.js
+++ b/api/app/clients/tools/dynamic/OpenAPIPlugin.spec.js
@@ -0,0 +1,72 @@
+const fs = require('fs');
+const { createOpenAPIPlugin, getSpec, readSpecFile } = require('./OpenAPIPlugin');
+
+global.fetch = jest.fn().mockImplementationOnce(() => {
+  return new Promise((resolve) => {
+    resolve({
+      ok: true,
+      json: () => Promise.resolve({ key: 'value' }),
+    });
+  });
+});
+jest.mock('fs', () => ({
+  promises: {
+    readFile: jest.fn(),
+  },
+  existsSync: jest.fn(),
+}));
+
+describe('readSpecFile', () => {
+  it('reads JSON file correctly', async () => {
+    fs.promises.readFile.mockResolvedValue(JSON.stringify({ test: 'value' }));
+    const result = await readSpecFile('test.json');
+    expect(result).toEqual({ test: 'value' });
+  });
+
+  it('reads YAML file correctly', async () => {
+    fs.promises.readFile.mockResolvedValue('test: value');
+    const result = await readSpecFile('test.yaml');
+    expect(result).toEqual({ test: 'value' });
+  });
+
+  it('handles error correctly', async () => {
+    fs.promises.readFile.mockRejectedValue(new Error('test error'));
+    const result = await readSpecFile('test.json');
+    expect(result).toBe(false);
+  });
+});
+
+describe('getSpec', () => {
+  it('fetches spec from url correctly', async () => {
+    const parsedJson = await getSpec('https://www.instacart.com/.well-known/ai-plugin.json');
+    const isObject = typeof parsedJson === 'object';
+    expect(isObject).toEqual(true);
+  });
+
+  it('reads spec from file correctly', async () => {
+    fs.existsSync.mockReturnValue(true);
+    fs.promises.readFile.mockResolvedValue(JSON.stringify({ test: 'value' }));
+    const result = await getSpec('test.json');
+    expect(result).toEqual({ test: 'value' });
+  });
+
+  it('returns false when file does not exist', async () => {
+    fs.existsSync.mockReturnValue(false);
+    const result = await getSpec('test.json');
+    expect(result).toBe(false);
+  });
+});
+
+describe('createOpenAPIPlugin', () => {
+  it('returns null when getSpec throws an error', async () => {
+    const result = await createOpenAPIPlugin({ data: { api: { url: 'invalid' } } });
+    expect(result).toBe(null);
+  });
+
+  it('returns null when no spec is found', async () => {
+    const result = await createOpenAPIPlugin({});
+    expect(result).toBe(null);
+  });
+
+  // Add more tests here for different scenarios
+});
--- a/api/app/clients/tools/index.js
+++ b/api/app/clients/tools/index.js
@@ -1,4 +1,4 @@
-const manifest = require('./manifest');
+const availableTools = require('./manifest.json');

 // Structured Tools
 const DALLE3 = require('./structured/DALLE3');
@@ -13,8 +13,23 @@ const TraversaalSearch = require('./structured/TraversaalSearch');
 const createOpenAIImageTools = require('./structured/OpenAIImageTools');
 const TavilySearchResults = require('./structured/TavilySearchResults');

+/** @type {Record<string, TPlugin | undefined>} */
+const manifestToolMap = {};
+
+/** @type {Array<TPlugin>} */
+const toolkits = [];
+
+availableTools.forEach((tool) => {
+  manifestToolMap[tool.pluginKey] = tool;
+  if (tool.toolkit === true) {
+    toolkits.push(tool);
+  }
+});
+
 module.exports = {
-  ...manifest,
+  toolkits,
+  availableTools,
+  manifestToolMap,
  // Structured Tools
  DALLE3,
  FluxAPI,
--- a/api/app/clients/tools/manifest.js
+++ b/api/app/clients/tools/manifest.js
@@ -1,20 +0,0 @@
-const availableTools = require('./manifest.json');
-
-/** @type {Record<string, TPlugin | undefined>} */
-const manifestToolMap = {};
-
-/** @type {Array<TPlugin>} */
-const toolkits = [];
-
-availableTools.forEach((tool) => {
-  manifestToolMap[tool.pluginKey] = tool;
-  if (tool.toolkit === true) {
-    toolkits.push(tool);
-  }
-});
-
-module.exports = {
-  toolkits,
-  availableTools,
-  manifestToolMap,
-};
--- a/api/app/clients/tools/manifest.json
+++ b/api/app/clients/tools/manifest.json
@@ -49,7 +49,7 @@
    "pluginKey": "image_gen_oai",
    "toolkit": true,
    "description": "Image Generation and Editing using OpenAI's latest state-of-the-art models",
-    "icon": "assets/image_gen_oai.png",
+    "icon": "/assets/image_gen_oai.png",
    "authConfig": [
      {
        "authField": "IMAGE_GEN_OAI_API_KEY",
@@ -75,7 +75,7 @@
    "name": "Browser",
    "pluginKey": "web-browser",
    "description": "Scrape and summarize webpage data",
-    "icon": "assets/web-browser.svg",
+    "icon": "/assets/web-browser.svg",
    "authConfig": [
      {
        "authField": "OPENAI_API_KEY",
@@ -170,7 +170,7 @@
    "name": "OpenWeather",
    "pluginKey": "open_weather",
    "description": "Get weather forecasts and historical data from the OpenWeather API",
-    "icon": "assets/openweather.png",
+    "icon": "/assets/openweather.png",
    "authConfig": [
      {
        "authField": "OPENWEATHER_API_KEY",
--- a/api/app/clients/tools/structured/AzureAISearch.js
+++ b/api/app/clients/tools/structured/AzureAISearch.js
@@ -1,7 +1,7 @@
 const { z } = require('zod');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
 const { SearchClient, AzureKeyCredential } = require('@azure/search-documents');
+const { logger } = require('~/config');

 class AzureAISearch extends Tool {
  // Constants for default values
--- a/api/app/clients/tools/structured/DALLE3.js
+++ b/api/app/clients/tools/structured/DALLE3.js
@@ -1,13 +1,14 @@
 const { z } = require('zod');
 const path = require('path');
 const OpenAI = require('openai');
+const fetch = require('node-fetch');
 const { v4: uuidv4 } = require('uuid');
-const { ProxyAgent, fetch } = require('undici');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
-const { getImageBasename } = require('@librechat/api');
+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');
+const { logger } = require('~/config');

 const displayMessage =
  "DALL-E displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
@@ -45,10 +46,7 @@ class DALLE3 extends Tool {
    }

    if (process.env.PROXY) {
-      const proxyAgent = new ProxyAgent(process.env.PROXY);
-      config.fetchOptions = {
-        dispatcher: proxyAgent,
-      };
+      config.httpAgent = new HttpsProxyAgent(process.env.PROXY);
    }

    /** @type {OpenAI} */
@@ -165,8 +163,7 @@ Error Message: ${error.message}`);
    if (this.isAgent) {
      let fetchOptions = {};
      if (process.env.PROXY) {
-        const proxyAgent = new ProxyAgent(process.env.PROXY);
-        fetchOptions.dispatcher = proxyAgent;
+        fetchOptions.agent = new HttpsProxyAgent(process.env.PROXY);
      }
      const imageResponse = await fetch(theImageUrl, fetchOptions);
      const arrayBuffer = await imageResponse.arrayBuffer();
--- a/api/app/clients/tools/structured/FluxAPI.js
+++ b/api/app/clients/tools/structured/FluxAPI.js
@@ -3,9 +3,9 @@ const axios = require('axios');
 const fetch = require('node-fetch');
 const { v4: uuidv4 } = require('uuid');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
 const { HttpsProxyAgent } = require('https-proxy-agent');
 const { FileContext, ContentTypes } = require('librechat-data-provider');
+const { logger } = require('~/config');

 const displayMessage =
  "Flux displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
--- a/api/app/clients/tools/structured/OpenAIImageTools.js
+++ b/api/app/clients/tools/structured/OpenAIImageTools.js
@@ -1,16 +1,67 @@
+const { z } = require('zod');
 const axios = require('axios');
 const { v4 } = require('uuid');
 const OpenAI = require('openai');
 const FormData = require('form-data');
-const { ProxyAgent } = require('undici');
 const { tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
 const { HttpsProxyAgent } = require('https-proxy-agent');
-const { logAxiosError, oaiToolkit } = require('@librechat/api');
 const { ContentTypes, EImageOutputType } = require('librechat-data-provider');
 const { getStrategyFunctions } = require('~/server/services/Files/strategies');
-const extractBaseURL = require('~/utils/extractBaseURL');
+const { logAxiosError, extractBaseURL } = require('~/utils');
 const { getFiles } = require('~/models/File');
+const { logger } = require('~/config');
+
+/** Default descriptions for image generation tool  */
+const DEFAULT_IMAGE_GEN_DESCRIPTION = `
+Generates high-quality, original images based solely on text, not using any uploaded reference images.
+
+When to use \`image_gen_oai\`:
+- To create entirely new images from detailed text descriptions that do NOT reference any image files.
+
+When NOT to use \`image_gen_oai\`:
+- If the user has uploaded any images and requests modifications, enhancements, or remixing based on those uploads → use \`image_edit_oai\` instead.
+
+Generated image IDs will be returned in the response, so you can refer to them in future requests made to \`image_edit_oai\`.
+`.trim();
+
+/** Default description for image editing tool  */
+const DEFAULT_IMAGE_EDIT_DESCRIPTION =
+  `Generates high-quality, original images based on text and one or more uploaded/referenced images.
+
+When to use \`image_edit_oai\`:
+- The user wants to modify, extend, or remix one **or more** uploaded images, either:
+- Previously generated, or in the current request (both to be included in the \`image_ids\` array).
+- Always when the user refers to uploaded images for editing, enhancement, remixing, style transfer, or combining elements.
+- Any current or existing images are to be used as visual guides.
+- If there are any files in the current request, they are more likely than not expected as references for image edit requests.
+
+When NOT to use \`image_edit_oai\`:
+- Brand-new generations that do not rely on an existing image → use \`image_gen_oai\` instead.
+
+Both generated and referenced image IDs will be returned in the response, so you can refer to them in future requests made to \`image_edit_oai\`.
+`.trim();
+
+/** Default prompt descriptions  */
+const DEFAULT_IMAGE_GEN_PROMPT_DESCRIPTION = `Describe the image you want in detail. 
+      Be highly specific—break your idea into layers: 
+      (1) main concept and subject,
+      (2) composition and position,
+      (3) lighting and mood,
+      (4) style, medium, or camera details,
+      (5) important features (age, expression, clothing, etc.),
+      (6) background.
+      Use positive, descriptive language and specify what should be included, not what to avoid. 
+      List number and characteristics of people/objects, and mention style/technical requirements (e.g., "DSLR photo, 85mm lens, golden hour").
+      Do not reference any uploaded images—use for new image creation from text only.`;
+
+const DEFAULT_IMAGE_EDIT_PROMPT_DESCRIPTION = `Describe the changes, enhancements, or new ideas to apply to the uploaded image(s).
+      Be highly specific—break your request into layers: 
+      (1) main concept or transformation,
+      (2) specific edits/replacements or composition guidance,
+      (3) desired style, mood, or technique,
+      (4) features/items to keep, change, or add (such as objects, people, clothing, lighting, etc.).
+      Use positive, descriptive language and clarify what should be included or changed, not what to avoid.
+      Always base this prompt on the most recently uploaded reference images.`;

 const displayMessage =
  "The tool displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
@@ -39,11 +90,21 @@ function returnValue(value) {
  return value;
 }

-function createAbortHandler() {
-  return function () {
-    logger.debug('[ImageGenOAI] Image generation aborted');
-  };
-}
+const getImageGenDescription = () => {
+  return process.env.IMAGE_GEN_OAI_DESCRIPTION || DEFAULT_IMAGE_GEN_DESCRIPTION;
+};
+
+const getImageEditDescription = () => {
+  return process.env.IMAGE_EDIT_OAI_DESCRIPTION || DEFAULT_IMAGE_EDIT_DESCRIPTION;
+};
+
+const getImageGenPromptDescription = () => {
+  return process.env.IMAGE_GEN_OAI_PROMPT_DESCRIPTION || DEFAULT_IMAGE_GEN_PROMPT_DESCRIPTION;
+};
+
+const getImageEditPromptDescription = () => {
+  return process.env.IMAGE_EDIT_OAI_PROMPT_DESCRIPTION || DEFAULT_IMAGE_EDIT_PROMPT_DESCRIPTION;
+};

 /**
 * Creates OpenAI Image tools (generation and editing)
@@ -53,9 +114,7 @@ function createAbortHandler() {
 * @param {string} fields.IMAGE_GEN_OAI_API_KEY - The OpenAI API key
 * @param {boolean} [fields.override] - Whether to override the API key check, necessary for app initialization
 * @param {MongoFile[]} [fields.imageFiles] - The images to be used for editing
- * @param {string} [fields.imageOutputType] - The image output type configuration
- * @param {string} [fields.fileStrategy] - The file storage strategy
- * @returns {Array<ReturnType<tool>>} - Array of image tools
+ * @returns {Array} - Array of image tools
 */
 function createOpenAIImageTools(fields = {}) {
  /** @type {boolean} Used to initialize the Tool without necessary variables. */
@@ -65,8 +124,8 @@ function createOpenAIImageTools(fields = {}) {
    throw new Error('This tool is only available for agents.');
  }
  const { req } = fields;
-  const imageOutputType = fields.imageOutputType || EImageOutputType.PNG;
-  const appFileStrategy = fields.fileStrategy;
+  const imageOutputType = req?.app.locals.imageOutputType || EImageOutputType.PNG;
+  const appFileStrategy = req?.app.locals.fileStrategy;

  const getApiKey = () => {
    const apiKey = process.env.IMAGE_GEN_OAI_API_KEY ?? '';
@@ -123,10 +182,7 @@ function createOpenAIImageTools(fields = {}) {
      }
      const clientConfig = { ...closureConfig };
      if (process.env.PROXY) {
-        const proxyAgent = new ProxyAgent(process.env.PROXY);
-        clientConfig.fetchOptions = {
-          dispatcher: proxyAgent,
-        };
+        clientConfig.httpAgent = new HttpsProxyAgent(process.env.PROXY);
      }

      /** @type {OpenAI} */
@@ -144,18 +200,10 @@ function createOpenAIImageTools(fields = {}) {
      }

      let resp;
-      /** @type {AbortSignal} */
-      let derivedSignal = null;
-      /** @type {() => void} */
-      let abortHandler = null;
-
      try {
-        if (runnableConfig?.signal) {
-          derivedSignal = AbortSignal.any([runnableConfig.signal]);
-          abortHandler = createAbortHandler();
-          derivedSignal.addEventListener('abort', abortHandler, { once: true });
-        }
-
+        const derivedSignal = runnableConfig?.signal
+          ? AbortSignal.any([runnableConfig.signal])
+          : undefined;
        resp = await openai.images.generate(
          {
            model: 'gpt-image-1',
@@ -179,10 +227,6 @@ 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) {
@@ -219,7 +263,46 @@ Error Message: ${error.message}`);
      ];
      return [response, { content, file_ids }];
    },
-    oaiToolkit.image_gen_oai,
+    {
+      name: 'image_gen_oai',
+      description: getImageGenDescription(),
+      schema: z.object({
+        prompt: z.string().max(32000).describe(getImageGenPromptDescription()),
+        background: z
+          .enum(['transparent', 'opaque', 'auto'])
+          .optional()
+          .describe(
+            'Sets transparency for the background. Must be one of transparent, opaque or auto (default). When transparent, the output format should be png or webp.',
+          ),
+        /*
+        n: z
+          .number()
+          .int()
+          .min(1)
+          .max(10)
+          .optional()
+          .describe('The number of images to generate. Must be between 1 and 10.'),
+        output_compression: z
+          .number()
+          .int()
+          .min(0)
+          .max(100)
+          .optional()
+          .describe('The compression level (0-100%) for webp or jpeg formats. Defaults to 100.'),
+           */
+        quality: z
+          .enum(['auto', 'high', 'medium', 'low'])
+          .optional()
+          .describe('The quality of the image. One of auto (default), high, medium, or low.'),
+        size: z
+          .enum(['auto', '1024x1024', '1536x1024', '1024x1536'])
+          .optional()
+          .describe(
+            'The size of the generated image. One of 1024x1024, 1536x1024 (landscape), 1024x1536 (portrait), or auto (default).',
+          ),
+      }),
+      responseFormat: 'content_and_artifact',
+    },
  );

  /**
@@ -233,10 +316,7 @@ Error Message: ${error.message}`);

      const clientConfig = { ...closureConfig };
      if (process.env.PROXY) {
-        const proxyAgent = new ProxyAgent(process.env.PROXY);
-        clientConfig.fetchOptions = {
-          dispatcher: proxyAgent,
-        };
+        clientConfig.httpAgent = new HttpsProxyAgent(process.env.PROXY);
      }

      const formData = new FormData();
@@ -328,17 +408,10 @@ Error Message: ${error.message}`);
        headers['Authorization'] = `Bearer ${apiKey}`;
      }

-      /** @type {AbortSignal} */
-      let derivedSignal = null;
-      /** @type {() => void} */
-      let abortHandler = null;
-
      try {
-        if (runnableConfig?.signal) {
-          derivedSignal = AbortSignal.any([runnableConfig.signal]);
-          abortHandler = createAbortHandler();
-          derivedSignal.addEventListener('abort', abortHandler, { once: true });
-        }
+        const derivedSignal = runnableConfig?.signal
+          ? AbortSignal.any([runnableConfig.signal])
+          : undefined;

        /** @type {import('axios').AxiosRequestConfig} */
        const axiosConfig = {
@@ -348,10 +421,6 @@ Error Message: ${error.message}`);
          baseURL,
        };

-        if (process.env.PROXY) {
-          axiosConfig.httpsAgent = new HttpsProxyAgent(process.env.PROXY);
-        }
-
        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,
@@ -397,13 +466,50 @@ 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);
-        }
      }
    },
-    oaiToolkit.image_edit_oai,
+    {
+      name: 'image_edit_oai',
+      description: getImageEditDescription(),
+      schema: z.object({
+        image_ids: z
+          .array(z.string())
+          .min(1)
+          .describe(
+            `
+IDs (image ID strings) of previously generated or uploaded images that should guide the edit.
+
+Guidelines:
+- If the user's request depends on any prior image(s), copy their image IDs into the \`image_ids\` array (in the same order the user refers to them).  
+- Never invent or hallucinate IDs; only use IDs that are still visible in the conversation context.
+- If no earlier image is relevant, omit the field entirely.
+`.trim(),
+          ),
+        prompt: z.string().max(32000).describe(getImageEditPromptDescription()),
+        /*
+        n: z
+          .number()
+          .int()
+          .min(1)
+          .max(10)
+          .optional()
+          .describe('The number of images to generate. Must be between 1 and 10. Defaults to 1.'),
+        */
+        quality: z
+          .enum(['auto', 'high', 'medium', 'low'])
+          .optional()
+          .describe(
+            'The quality of the image. One of auto (default), high, medium, or low. High/medium/low only supported for gpt-image-1.',
+          ),
+        size: z
+          .enum(['auto', '1024x1024', '1536x1024', '1024x1536', '256x256', '512x512'])
+          .optional()
+          .describe(
+            'The size of the generated images. For gpt-image-1: auto (default), 1024x1024, 1536x1024, 1024x1536. For dall-e-2: 256x256, 512x512, 1024x1024.',
+          ),
+      }),
+      responseFormat: 'content_and_artifact',
+    },
  );

  return [imageGenTool, imageEditTool];
--- a/api/app/clients/tools/structured/OpenWeather.js
+++ b/api/app/clients/tools/structured/OpenWeather.js
@@ -232,7 +232,7 @@ class OpenWeather extends Tool {

      if (['current_forecast', 'timestamp', 'daily_aggregation', 'overview'].includes(action)) {
        if (typeof finalLat !== 'number' || typeof finalLon !== 'number') {
-          return 'Error: lat and lon are required and must be numbers for this action (or specify \'city\').';
+          return "Error: lat and lon are required and must be numbers for this action (or specify 'city').";
        }
      }

@@ -243,7 +243,7 @@ class OpenWeather extends Tool {
      let dt;
      if (action === 'timestamp') {
        if (!date) {
-          return 'Error: For timestamp action, a \'date\' in YYYY-MM-DD format is required.';
+          return "Error: For timestamp action, a 'date' in YYYY-MM-DD format is required.";
        }
        dt = this.convertDateToUnix(date);
      }
--- a/api/app/clients/tools/structured/StableDiffusion.js
+++ b/api/app/clients/tools/structured/StableDiffusion.js
@@ -6,9 +6,9 @@ const axios = require('axios');
 const sharp = require('sharp');
 const { v4: uuidv4 } = require('uuid');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
 const { FileContext, ContentTypes } = require('librechat-data-provider');
 const paths = require('~/config/paths');
+const { logger } = require('~/config');

 const displayMessage =
  "Stable Diffusion displayed an image. All generated images are already plainly visible, so don't repeat the descriptions in detail. Do not list download links as they are available in the UI already. The user may download the images by clicking on them, but do not mention anything about downloading to the user.";
@@ -18,7 +18,7 @@ class StableDiffusionAPI extends Tool {
    super();
    /** @type {string} User ID */
    this.userId = fields.userId;
-    /** @type {ServerRequest | undefined} Express Request object, only provided by ToolService */
+    /** @type {Express.Request | undefined} Express Request object, only provided by ToolService */
    this.req = fields.req;
    /** @type {boolean} Used to initialize the Tool without necessary variables. */
    this.override = fields.override ?? false;
--- a/api/app/clients/tools/structured/TraversaalSearch.js
+++ b/api/app/clients/tools/structured/TraversaalSearch.js
@@ -1,7 +1,7 @@
 const { z } = require('zod');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
 const { getEnvironmentVariable } = require('@langchain/core/utils/env');
+const { logger } = require('~/config');

 /**
 * Tool for the Traversaal AI search API, Ares.
--- a/api/app/clients/tools/structured/Wolfram.js
+++ b/api/app/clients/tools/structured/Wolfram.js
@@ -1,8 +1,8 @@
 /* eslint-disable no-useless-escape */
-const { z } = require('zod');
 const axios = require('axios');
+const { z } = require('zod');
 const { Tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
+const { logger } = require('~/config');

 class WolframAlphaAPI extends Tool {
  constructor(fields) {
--- a/api/app/clients/tools/structured/YouTube.js
+++ b/api/app/clients/tools/structured/YouTube.js
@@ -1,9 +1,9 @@
-const { ytToolkit } = require('@librechat/api');
+const { z } = require('zod');
 const { tool } = require('@langchain/core/tools');
 const { youtube } = require('@googleapis/youtube');
-const { logger } = require('@librechat/data-schemas');
 const { YoutubeTranscript } = require('youtube-transcript');
 const { getApiKey } = require('./credentials');
+const { logger } = require('~/config');

 function extractVideoId(url) {
  const rawIdRegex = /^[a-zA-Z0-9_-]{11}$/;
@@ -42,94 +42,160 @@ function createYouTubeTools(fields = {}) {
    auth: apiKey,
  });

-  const searchTool = tool(async ({ query, maxResults = 5 }) => {
-    const response = await youtubeClient.search.list({
-      part: 'snippet',
-      q: query,
-      type: 'video',
-      maxResults: maxResults || 5,
-    });
-    const result = response.data.items.map((item) => ({
-      title: item.snippet.title,
-      description: item.snippet.description,
-      url: `https://www.youtube.com/watch?v=${item.id.videoId}`,
-    }));
-    return JSON.stringify(result, null, 2);
-  }, ytToolkit.youtube_search);
+  const searchTool = tool(
+    async ({ query, maxResults = 5 }) => {
+      const response = await youtubeClient.search.list({
+        part: 'snippet',
+        q: query,
+        type: 'video',
+        maxResults: maxResults || 5,
+      });
+      const result = response.data.items.map((item) => ({
+        title: item.snippet.title,
+        description: item.snippet.description,
+        url: `https://www.youtube.com/watch?v=${item.id.videoId}`,
+      }));
+      return JSON.stringify(result, null, 2);
+    },
+    {
+      name: 'youtube_search',
+      description: `Search for YouTube videos by keyword or phrase.
+- Required: query (search terms to find videos)
+- Optional: maxResults (number of videos to return, 1-50, default: 5)
+- Returns: List of videos with titles, descriptions, and URLs
+- Use for: Finding specific videos, exploring content, research
+Example: query="cooking pasta tutorials" maxResults=3`,
+      schema: z.object({
+        query: z.string().describe('Search query terms'),
+        maxResults: z.number().int().min(1).max(50).optional().describe('Number of results (1-50)'),
+      }),
+    },
+  );

-  const infoTool = tool(async ({ url }) => {
-    const videoId = extractVideoId(url);
-    if (!videoId) {
-      throw new Error('Invalid YouTube URL or video ID');
-    }
+  const infoTool = tool(
+    async ({ url }) => {
+      const videoId = extractVideoId(url);
+      if (!videoId) {
+        throw new Error('Invalid YouTube URL or video ID');
+      }

-    const response = await youtubeClient.videos.list({
-      part: 'snippet,statistics',
-      id: videoId,
-    });
+      const response = await youtubeClient.videos.list({
+        part: 'snippet,statistics',
+        id: videoId,
+      });

-    if (!response.data.items?.length) {
-      throw new Error('Video not found');
-    }
-    const video = response.data.items[0];
+      if (!response.data.items?.length) {
+        throw new Error('Video not found');
+      }
+      const video = response.data.items[0];

-    const result = {
-      title: video.snippet.title,
-      description: video.snippet.description,
-      views: video.statistics.viewCount,
-      likes: video.statistics.likeCount,
-      comments: video.statistics.commentCount,
-    };
-    return JSON.stringify(result, null, 2);
-  }, ytToolkit.youtube_info);
+      const result = {
+        title: video.snippet.title,
+        description: video.snippet.description,
+        views: video.statistics.viewCount,
+        likes: video.statistics.likeCount,
+        comments: video.statistics.commentCount,
+      };
+      return JSON.stringify(result, null, 2);
+    },
+    {
+      name: 'youtube_info',
+      description: `Get detailed metadata and statistics for a specific YouTube video.
+- Required: url (full YouTube URL or video ID)
+- Returns: Video title, description, view count, like count, comment count
+- Use for: Getting video metrics and basic metadata
+- DO NOT USE FOR VIDEO SUMMARIES, USE TRANSCRIPTS FOR COMPREHENSIVE ANALYSIS
+- Accepts both full URLs and video IDs
+Example: url="https://youtube.com/watch?v=abc123" or url="abc123"`,
+      schema: z.object({
+        url: z.string().describe('YouTube video URL or ID'),
+      }),
+    },
+  );

-  const commentsTool = tool(async ({ url, maxResults = 10 }) => {
-    const videoId = extractVideoId(url);
-    if (!videoId) {
-      throw new Error('Invalid YouTube URL or video ID');
-    }
+  const commentsTool = tool(
+    async ({ url, maxResults = 10 }) => {
+      const videoId = extractVideoId(url);
+      if (!videoId) {
+        throw new Error('Invalid YouTube URL or video ID');
+      }

-    const response = await youtubeClient.commentThreads.list({
-      part: 'snippet',
-      videoId,
-      maxResults: maxResults || 10,
-    });
+      const response = await youtubeClient.commentThreads.list({
+        part: 'snippet',
+        videoId,
+        maxResults: maxResults || 10,
+      });

-    const result = response.data.items.map((item) => ({
-      author: item.snippet.topLevelComment.snippet.authorDisplayName,
-      text: item.snippet.topLevelComment.snippet.textDisplay,
-      likes: item.snippet.topLevelComment.snippet.likeCount,
-    }));
-    return JSON.stringify(result, null, 2);
-  }, ytToolkit.youtube_comments);
+      const result = response.data.items.map((item) => ({
+        author: item.snippet.topLevelComment.snippet.authorDisplayName,
+        text: item.snippet.topLevelComment.snippet.textDisplay,
+        likes: item.snippet.topLevelComment.snippet.likeCount,
+      }));
+      return JSON.stringify(result, null, 2);
+    },
+    {
+      name: 'youtube_comments',
+      description: `Retrieve top-level comments from a YouTube video.
+- Required: url (full YouTube URL or video ID)
+- Optional: maxResults (number of comments, 1-50, default: 10)
+- Returns: Comment text, author names, like counts
+- Use for: Sentiment analysis, audience feedback, engagement review
+Example: url="abc123" maxResults=20`,
+      schema: z.object({
+        url: z.string().describe('YouTube video URL or ID'),
+        maxResults: z
+          .number()
+          .int()
+          .min(1)
+          .max(50)
+          .optional()
+          .describe('Number of comments to retrieve'),
+      }),
+    },
+  );

-  const transcriptTool = tool(async ({ url }) => {
-    const videoId = extractVideoId(url);
-    if (!videoId) {
-      throw new Error('Invalid YouTube URL or video ID');
-    }
-
-    try {
-      try {
-        const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'en' });
-        return parseTranscript(transcript);
-      } catch (e) {
-        logger.error(e);
+  const transcriptTool = tool(
+    async ({ url }) => {
+      const videoId = extractVideoId(url);
+      if (!videoId) {
+        throw new Error('Invalid YouTube URL or video ID');
      }

      try {
-        const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'de' });
-        return parseTranscript(transcript);
-      } catch (e) {
-        logger.error(e);
-      }
+        try {
+          const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'en' });
+          return parseTranscript(transcript);
+        } catch (e) {
+          logger.error(e);
+        }

-      const transcript = await YoutubeTranscript.fetchTranscript(videoId);
-      return parseTranscript(transcript);
-    } catch (error) {
-      throw new Error(`Failed to fetch transcript: ${error.message}`);
-    }
-  }, ytToolkit.youtube_transcript);
+        try {
+          const transcript = await YoutubeTranscript.fetchTranscript(videoId, { lang: 'de' });
+          return parseTranscript(transcript);
+        } catch (e) {
+          logger.error(e);
+        }
+
+        const transcript = await YoutubeTranscript.fetchTranscript(videoId);
+        return parseTranscript(transcript);
+      } catch (error) {
+        throw new Error(`Failed to fetch transcript: ${error.message}`);
+      }
+    },
+    {
+      name: 'youtube_transcript',
+      description: `Fetch and parse the transcript/captions of a YouTube video.
+- Required: url (full YouTube URL or video ID)
+- Returns: Full video transcript as plain text
+- Use for: Content analysis, summarization, translation reference
+- This is the "Go-to" tool for analyzing actual video content
+- Attempts to fetch English first, then German, then any available language
+Example: url="https://youtube.com/watch?v=abc123"`,
+      schema: z.object({
+        url: z.string().describe('YouTube video URL or ID'),
+      }),
+    },
+  );

  return [searchTool, infoTool, commentsTool, transcriptTool];
 }
--- a/api/app/clients/tools/structured/specs/DALLE3-proxy.spec.js
+++ b/api/app/clients/tools/structured/specs/DALLE3-proxy.spec.js
@@ -1,60 +0,0 @@
-const DALLE3 = require('../DALLE3');
-const { ProxyAgent } = require('undici');
-
-jest.mock('tiktoken');
-const processFileURL = jest.fn();
-
-describe('DALLE3 Proxy Configuration', () => {
-  let originalEnv;
-
-  beforeAll(() => {
-    originalEnv = { ...process.env };
-  });
-
-  beforeEach(() => {
-    jest.resetModules();
-    process.env = { ...originalEnv };
-  });
-
-  afterEach(() => {
-    process.env = originalEnv;
-  });
-
-  it('should configure ProxyAgent in fetchOptions.dispatcher when PROXY env is set', () => {
-    // Set proxy environment variable
-    process.env.PROXY = 'http://proxy.example.com:8080';
-    process.env.DALLE_API_KEY = 'test-api-key';
-
-    // Create instance
-    const dalleWithProxy = new DALLE3({ processFileURL });
-
-    // Check that the openai client exists
-    expect(dalleWithProxy.openai).toBeDefined();
-
-    // Check that _options exists and has fetchOptions with a dispatcher
-    expect(dalleWithProxy.openai._options).toBeDefined();
-    expect(dalleWithProxy.openai._options.fetchOptions).toBeDefined();
-    expect(dalleWithProxy.openai._options.fetchOptions.dispatcher).toBeDefined();
-    expect(dalleWithProxy.openai._options.fetchOptions.dispatcher).toBeInstanceOf(ProxyAgent);
-  });
-
-  it('should not configure ProxyAgent when PROXY env is not set', () => {
-    // Ensure PROXY is not set
-    delete process.env.PROXY;
-    process.env.DALLE_API_KEY = 'test-api-key';
-
-    // Create instance
-    const dalleWithoutProxy = new DALLE3({ processFileURL });
-
-    // Check that the openai client exists
-    expect(dalleWithoutProxy.openai).toBeDefined();
-
-    // Check that _options exists but fetchOptions either doesn't exist or doesn't have a dispatcher
-    expect(dalleWithoutProxy.openai._options).toBeDefined();
-
-    // fetchOptions should either not exist or not have a dispatcher
-    if (dalleWithoutProxy.openai._options.fetchOptions) {
-      expect(dalleWithoutProxy.openai._options.fetchOptions.dispatcher).toBeUndefined();
-    }
-  });
-});
--- a/api/app/clients/tools/structured/specs/DALLE3.spec.js
+++ b/api/app/clients/tools/structured/specs/DALLE3.spec.js
@@ -1,30 +1,31 @@
 const OpenAI = require('openai');
-const { logger } = require('@librechat/data-schemas');
 const DALLE3 = require('../DALLE3');

-jest.mock('openai');
-jest.mock('@librechat/data-schemas', () => {
-  return {
-    logger: {
-      info: jest.fn(),
-      warn: jest.fn(),
-      debug: jest.fn(),
-      error: jest.fn(),
-    },
-  };
-});
+const { logger } = require('~/config');

-jest.mock('tiktoken', () => {
-  return {
-    encoding_for_model: jest.fn().mockReturnValue({
-      encode: jest.fn(),
-      decode: jest.fn(),
-    }),
-  };
-});
+jest.mock('openai');

 const processFileURL = jest.fn();

+jest.mock('~/server/services/Files/images', () => ({
+  getImageBasename: jest.fn().mockImplementation((url) => {
+    // Split the URL by '/'
+    const parts = url.split('/');
+
+    // Get the last part of the URL
+    const lastPart = parts.pop();
+
+    // Check if the last part of the URL matches the image extension regex
+    const imageExtensionRegex = /\.(jpg|jpeg|png|gif|bmp|tiff|svg)$/i;
+    if (imageExtensionRegex.test(lastPart)) {
+      return lastPart;
+    }
+
+    // If the regex test fails, return an empty string
+    return '';
+  }),
+}));
+
 const generate = jest.fn();
 OpenAI.mockImplementation(() => ({
  images: {
@@ -36,11 +37,6 @@ jest.mock('fs', () => {
  return {
    existsSync: jest.fn(),
    mkdirSync: jest.fn(),
-    promises: {
-      writeFile: jest.fn(),
-      readFile: jest.fn(),
-      unlink: jest.fn(),
-    },
  };
 });

--- a/api/app/clients/tools/util/fileSearch.js
+++ b/api/app/clients/tools/util/fileSearch.js
@@ -1,46 +1,26 @@
 const { z } = require('zod');
 const axios = require('axios');
 const { tool } = require('@langchain/core/tools');
-const { logger } = require('@librechat/data-schemas');
-const { generateShortLivedToken } = require('@librechat/api');
 const { Tools, EToolResources } = require('librechat-data-provider');
-const { filterFilesByAgentAccess } = require('~/server/services/Files/permissions');
 const { getFiles } = require('~/models/File');
+const { logger } = require('~/config');

 /**
 *
 * @param {Object} options
 * @param {ServerRequest} options.req
 * @param {Agent['tool_resources']} options.tool_resources
- * @param {string} [options.agentId] - The agent ID for file access control
 * @returns {Promise<{
 *   files: Array<{ file_id: string; filename: string }>,
 *   toolContext: string
 * }>}
 */
 const primeFiles = async (options) => {
-  const { tool_resources, req, agentId } = options;
+  const { tool_resources } = 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 ?? [];
-
-  // 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);
+  const dbFiles = ((await getFiles({ file_id: { $in: file_ids } })) ?? []).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.`;

@@ -68,19 +48,18 @@ const primeFiles = async (options) => {
 /**
 *
 * @param {Object} options
- * @param {string} options.userId
+ * @param {ServerRequest} options.req
 * @param {Array<{ file_id: string; filename: string }>} options.files
 * @param {string} [options.entity_id]
- * @param {boolean} [options.fileCitations=false] - Whether to include citation instructions
 * @returns
 */
-const createFileSearchTool = async ({ userId, files, entity_id, fileCitations = false }) => {
+const createFileSearchTool = async ({ req, files, entity_id }) => {
  return tool(
    async ({ query }) => {
      if (files.length === 0) {
        return 'No files to search. Instruct the user to add files for the search.';
      }
-      const jwtToken = generateShortLivedToken(userId);
+      const jwtToken = req.headers.authorization.split(' ')[1];
      if (!jwtToken) {
        return 'There was an error authenticating the file search request.';
      }
@@ -126,13 +105,11 @@ const createFileSearchTool = async ({ userId, files, entity_id, fileCitations =
      }

      const formattedResults = validResults
-        .flatMap((result, fileIndex) =>
+        .flatMap((result) =>
          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
@@ -142,41 +119,18 @@ const createFileSearchTool = async ({ userId, files, entity_id, fileCitations =

      const formattedString = formattedResults
        .map(
-          (result, index) =>
-            `File: ${result.filename}${
-              fileCitations ? `\nAnchor: \\ue202turn0file${index} (${result.filename})` : ''
-            }\nRelevance: ${(1.0 - result.distance).toFixed(4)}\nContent: ${result.content}\n`,
+          (result) =>
+            `File: ${result.filename}\nRelevance: ${1.0 - result.distance.toFixed(4)}\nContent: ${
+              result.content
+            }\n`,
        )
        .join('\n---\n');

-      const sources = formattedResults.map((result) => ({
-        type: 'file',
-        fileId: result.file_id,
-        content: result.content,
-        fileName: result.filename,
-        relevance: 1.0 - result.distance,
-        pages: result.page ? [result.page] : [],
-        pageRelevance: result.page ? { [result.page]: 1.0 - result.distance } : {},
-      }));
-
-      return [formattedString, { [Tools.file_search]: { sources, fileCitations } }];
+      return formattedString;
    },
    {
      name: Tools.file_search,
-      responseFormat: 'content_and_artifact',
-      description: `Performs semantic search across attached "${Tools.file_search}" documents using natural language queries. This tool analyzes the content of uploaded files to find relevant information, quotes, and passages that best match your query. Use this to extract specific information or find relevant sections within the available documents.${
-        fileCitations
-          ? `
-
-**CITE FILE SEARCH RESULTS:**
-Use anchor markers immediately after statements derived from file content. Reference the filename in your text:
- File citation: "The document.pdf states that... \\ue202turn0file0"  
- Page reference: "According to report.docx... \\ue202turn0file1"
- Multi-file: "Multiple sources confirm... \\ue200\\ue202turn0file0\\ue202turn0file1\\ue201"
-
-**ALWAYS mention the filename in your text before the citation marker. NEVER use markdown links or footnotes.**`
-          : ''
-      }`,
+      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.`,
      schema: z.object({
        query: z
          .string()
--- a/api/app/clients/tools/util/handleOpenAIErrors.js
+++ b/api/app/clients/tools/util/handleOpenAIErrors.js
@@ -1,5 +1,5 @@
 const OpenAI = require('openai');
-const { logger } = require('@librechat/data-schemas');
+const { logger } = require('~/config');

 /**
 * Handles errors that may occur when making requests to OpenAI's API.
--- a/api/app/clients/tools/util/handleTools.js
+++ b/api/app/clients/tools/util/handleTools.js
@@ -1,21 +1,14 @@
-const { logger } = require('@librechat/data-schemas');
 const { SerpAPI } = require('@langchain/community/tools/serpapi');
 const { Calculator } = require('@langchain/community/tools/calculator');
 const { EnvVar, createCodeExecutionTool, createSearchTool } = require('@librechat/agents');
-const {
-  checkAccess,
-  createSafeUser,
-  mcpToolPattern,
-  loadWebSearchAuth,
-} = require('@librechat/api');
 const {
  Tools,
  Constants,
-  Permissions,
  EToolResources,
-  PermissionTypes,
+  loadWebSearchAuth,
  replaceSpecialVars,
 } = require('librechat-data-provider');
+const { getUserPluginAuthValue } = require('~/server/services/PluginService');
 const {
  availableTools,
  manifestToolMap,
@@ -35,11 +28,11 @@ const {
 } = require('../');
 const { primeFiles: primeCodeFiles } = require('~/server/services/Files/Code/process');
 const { createFileSearchTool, primeFiles: primeSearchFiles } = require('./fileSearch');
-const { getUserPluginAuthValue } = require('~/server/services/PluginService');
-const { createMCPTool, createMCPTools } = require('~/server/services/MCP');
 const { loadAuthValues } = require('~/server/services/Tools/credentials');
-const { getMCPServerTools } = require('~/server/services/Config');
-const { getRoleByName } = require('~/models/Role');
+const { createMCPTool } = require('~/server/services/MCP');
+const { logger } = require('~/config');
+
+const mcpToolPattern = new RegExp(`^.+${Constants.mcp_delimiter}.+$`);

 /**
 * Validates the availability and authentication of tools for a user based on environment variables or user-specific plugin authentication values.
@@ -100,7 +93,7 @@ const validateTools = async (user, tools = []) => {
    return Array.from(validToolsSet.values());
  } catch (err) {
    logger.error('[validateTools] There was a problem validating tools', err);
-    throw new Error(err);
+    throw new Error('There was a problem validating tools');
  }
 };

@@ -134,37 +127,27 @@ const getAuthFields = (toolKey) => {

 /**
 *
- * @param {object} params
- * @param {string} params.user
- * @param {Record<string, Record<string, string>>} [object.userMCPAuthMap]
- * @param {AbortSignal} [object.signal]
- * @param {Pick<Agent, 'id' | 'provider' | 'model'>} [params.agent]
- * @param {string} [params.model]
- * @param {EModelEndpoint} [params.endpoint]
- * @param {LoadToolOptions} [params.options]
- * @param {boolean} [params.useSpecs]
- * @param {Array<string>} params.tools
- * @param {boolean} [params.functions]
- * @param {boolean} [params.returnMap]
- * @param {AppConfig['webSearch']} [params.webSearch]
- * @param {AppConfig['fileStrategy']} [params.fileStrategy]
- * @param {AppConfig['imageOutputType']} [params.imageOutputType]
+ * @param {object} object
+ * @param {string} object.user
+ * @param {Pick<Agent, 'id' | 'provider' | 'model'>} [object.agent]
+ * @param {string} [object.model]
+ * @param {EModelEndpoint} [object.endpoint]
+ * @param {LoadToolOptions} [object.options]
+ * @param {boolean} [object.useSpecs]
+ * @param {Array<string>} object.tools
+ * @param {boolean} [object.functions]
+ * @param {boolean} [object.returnMap]
 * @returns {Promise<{ loadedTools: Tool[], toolContextMap: Object<string, any> } | Record<string,Tool>>}
 */
 const loadTools = async ({
  user,
  agent,
  model,
-  signal,
  endpoint,
-  userMCPAuthMap,
  tools = [],
  options = {},
  functions = true,
  returnMap = false,
-  webSearch,
-  fileStrategy,
-  imageOutputType,
 }) => {
  const toolConstructors = {
    flux: FluxAPI,
@@ -223,8 +206,6 @@ const loadTools = async ({
        ...authValues,
        isAgent: !!agent,
        req: options.req,
-        imageOutputType,
-        fileStrategy,
        imageFiles,
      });
    },
@@ -240,7 +221,7 @@ const loadTools = async ({
  const imageGenOptions = {
    isAgent: !!agent,
    req: options.req,
-    fileStrategy,
+    fileStrategy: options.fileStrategy,
    processFileURL: options.processFileURL,
    returnMetadata: options.returnMetadata,
    uploadImageBuffer: options.uploadImageBuffer,
@@ -255,7 +236,7 @@ const loadTools = async ({

  /** @type {Record<string, string>} */
  const toolContextMap = {};
-  const requestedMCPTools = {};
+  const appTools = options.req?.app?.locals?.availableTools ?? {};

  for (const tool of tools) {
    if (tool === Tools.execute_code) {
@@ -265,13 +246,7 @@ const loadTools = async ({
          authFields: [EnvVar.CODE_API_KEY],
        });
        const codeApiKey = authValues[EnvVar.CODE_API_KEY];
-        const { files, toolContext } = await primeCodeFiles(
-          {
-            ...options,
-            agentId: agent?.id,
-          },
-          codeApiKey,
-        );
+        const { files, toolContext } = await primeCodeFiles(options, codeApiKey);
        if (toolContext) {
          toolContextMap[tool] = toolContext;
        }
@@ -286,43 +261,19 @@ const loadTools = async ({
      continue;
    } else if (tool === Tools.file_search) {
      requestedTools[tool] = async () => {
-        const { files, toolContext } = await primeSearchFiles({
-          ...options,
-          agentId: agent?.id,
-        });
+        const { files, toolContext } = await primeSearchFiles(options);
        if (toolContext) {
          toolContextMap[tool] = toolContext;
        }
-
-        /** @type {boolean | undefined} Check if user has FILE_CITATIONS permission */
-        let fileCitations;
-        if (fileCitations == null && options.req?.user != null) {
-          try {
-            fileCitations = await checkAccess({
-              user: options.req.user,
-              permissionType: PermissionTypes.FILE_CITATIONS,
-              permissions: [Permissions.USE],
-              getRoleByName,
-            });
-          } catch (error) {
-            logger.error('[handleTools] FILE_CITATIONS permission check failed:', error);
-            fileCitations = false;
-          }
-        }
-
-        return createFileSearchTool({
-          userId: user,
-          files,
-          entity_id: agent?.id,
-          fileCitations,
-        });
+        return createFileSearchTool({ req: options.req, files, entity_id: agent?.id });
      };
      continue;
    } else if (tool === Tools.web_search) {
+      const webSearchConfig = options?.req?.app?.locals?.webSearch;
      const result = await loadWebSearchAuth({
        userId: user,
        loadAuthValues,
-        webSearchConfig: webSearch,
+        webSearchConfig,
      });
      const { onSearchResults, onGetHighlights } = options?.[Tools.web_search] ?? {};
      requestedTools[tool] = async () => {
@@ -344,34 +295,14 @@ Current Date & Time: ${replaceSpecialVars({ text: '{{iso_datetime}}' })}
        });
      };
      continue;
-    } else if (tool && mcpToolPattern.test(tool)) {
-      const [toolName, serverName] = tool.split(Constants.mcp_delimiter);
-      if (toolName === Constants.mcp_server) {
-        /** Placeholder used for UI purposes */
-        continue;
-      }
-      if (serverName && options.req?.config?.mcpConfig?.[serverName] == null) {
-        logger.warn(
-          `MCP server "${serverName}" for "${toolName}" tool is not configured${agent?.id != null && agent.id ? ` but attached to "${agent.id}"` : ''}`,
-        );
-        continue;
-      }
-      if (toolName === Constants.mcp_all) {
-        requestedMCPTools[serverName] = [
-          {
-            type: 'all',
-            serverName,
-          },
-        ];
-        continue;
-      }
-
-      requestedMCPTools[serverName] = requestedMCPTools[serverName] || [];
-      requestedMCPTools[serverName].push({
-        type: 'single',
-        toolKey: tool,
-        serverName,
-      });
+    } else if (tool && appTools[tool] && mcpToolPattern.test(tool)) {
+      requestedTools[tool] = async () =>
+        createMCPTool({
+          req: options.req,
+          toolKey: tool,
+          model: agent?.model ?? model,
+          provider: agent?.provider ?? endpoint,
+        });
      continue;
    }

@@ -411,75 +342,6 @@ Current Date & Time: ${replaceSpecialVars({ text: '{{iso_datetime}}' })}
  }

  const loadedTools = (await Promise.all(toolPromises)).flatMap((plugin) => plugin || []);
-  const mcpToolPromises = [];
-  /** MCP server tools are initialized sequentially by server */
-  let index = -1;
-  const failedMCPServers = new Set();
-  const safeUser = createSafeUser(options.req?.user);
-  for (const [serverName, toolConfigs] of Object.entries(requestedMCPTools)) {
-    index++;
-    /** @type {LCAvailableTools} */
-    let availableTools;
-    for (const config of toolConfigs) {
-      try {
-        if (failedMCPServers.has(serverName)) {
-          continue;
-        }
-        const mcpParams = {
-          index,
-          signal,
-          user: safeUser,
-          userMCPAuthMap,
-          res: options.res,
-          model: agent?.model ?? model,
-          serverName: config.serverName,
-          provider: agent?.provider ?? endpoint,
-        };
-
-        if (config.type === 'all' && toolConfigs.length === 1) {
-          /** Handle async loading for single 'all' tool config */
-          mcpToolPromises.push(
-            createMCPTools(mcpParams).catch((error) => {
-              logger.error(`Error loading ${serverName} tools:`, error);
-              return null;
-            }),
-          );
-          continue;
-        }
-        if (!availableTools) {
-          try {
-            availableTools = await getMCPServerTools(serverName);
-          } catch (error) {
-            logger.error(`Error fetching available tools for MCP server ${serverName}:`, error);
-          }
-        }
-
-        /** Handle synchronous loading */
-        const mcpTool =
-          config.type === 'all'
-            ? await createMCPTools(mcpParams)
-            : await createMCPTool({
-                ...mcpParams,
-                availableTools,
-                toolKey: config.toolKey,
-              });
-
-        if (Array.isArray(mcpTool)) {
-          loadedTools.push(...mcpTool);
-        } else if (mcpTool) {
-          loadedTools.push(mcpTool);
-        } else {
-          failedMCPServers.add(serverName);
-          logger.warn(
-            `MCP tool creation failed for "${config.toolKey}", server may be unavailable or unauthenticated.`,
-          );
-        }
-      } catch (error) {
-        logger.error(`Error loading MCP tool for server ${serverName}:`, error);
-      }
-    }
-  }
-  loadedTools.push(...(await Promise.all(mcpToolPromises)).flatMap((plugin) => plugin || []));
  return { loadedTools, toolContextMap };
 };

--- a/api/app/clients/tools/util/handleTools.test.js
+++ b/api/app/clients/tools/util/handleTools.test.js
@@ -1,5 +1,8 @@
-const mongoose = require('mongoose');
-const { MongoMemoryServer } = require('mongodb-memory-server');
+const mockUser = {
+  _id: 'fakeId',
+  save: jest.fn(),
+  findByIdAndDelete: jest.fn(),
+};

 const mockPluginService = {
  updateUserPluginAuth: jest.fn(),
@@ -7,38 +10,23 @@ const mockPluginService = {
  getUserPluginAuthValue: jest.fn(),
 };

+jest.mock('~/models/User', () => {
+  return function () {
+    return mockUser;
+  };
+});
+
 jest.mock('~/server/services/PluginService', () => mockPluginService);

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

-const { User } = require('~/db/models');
+const User = require('~/models/User');
 const PluginService = require('~/server/services/PluginService');
 const { validateTools, loadTools, loadToolWithAuth } = require('./handleTools');
 const { StructuredSD, availableTools, DALLE3 } = require('../');

 describe('Tool Handlers', () => {
-  let mongoServer;
  let fakeUser;
  const pluginKey = 'dalle';
  const pluginKey2 = 'wolfram';
@@ -49,9 +37,7 @@ describe('Tool Handlers', () => {
  const authConfigs = mainPlugin.authConfig;

  beforeAll(async () => {
-    mongoServer = await MongoMemoryServer.create();
-    const mongoUri = mongoServer.getUri();
-    await mongoose.connect(mongoUri);
+    mockUser.save.mockResolvedValue(undefined);

    const userAuthValues = {};
    mockPluginService.getUserPluginAuthValue.mockImplementation((userId, authField) => {
@@ -92,36 +78,9 @@ describe('Tool Handlers', () => {
  });

  afterAll(async () => {
-    await mongoose.disconnect();
-    await mongoServer.stop();
-  });
-
-  beforeEach(async () => {
-    // Clear mocks but not the database since we need the user to persist
-    jest.clearAllMocks();
-
-    // Reset the mock implementations
-    const userAuthValues = {};
-    mockPluginService.getUserPluginAuthValue.mockImplementation((userId, authField) => {
-      return userAuthValues[`${userId}-${authField}`];
-    });
-    mockPluginService.updateUserPluginAuth.mockImplementation(
-      (userId, authField, _pluginKey, credential) => {
-        const fields = authField.split('||');
-        fields.forEach((field) => {
-          userAuthValues[`${userId}-${field}`] = credential;
-        });
-      },
-    );
-
-    // Re-add the auth configs for the user
+    await mockUser.findByIdAndDelete(fakeUser._id);
    for (const authConfig of authConfigs) {
-      await PluginService.updateUserPluginAuth(
-        fakeUser._id,
-        authConfig.authField,
-        pluginKey,
-        mockCredential,
-      );
+      await PluginService.deleteUserPluginAuth(fakeUser._id, authConfig.authField);
    }
  });

@@ -171,6 +130,7 @@ describe('Tool Handlers', () => {
    beforeAll(async () => {
      const toolMap = await loadTools({
        user: fakeUser._id,
+        model: BaseLLM,
        tools: sampleTools,
        returnMap: true,
        useSpecs: true,
@@ -264,6 +224,7 @@ describe('Tool Handlers', () => {
    it('returns an empty object when no tools are requested', async () => {
      toolFunctions = await loadTools({
        user: fakeUser._id,
+        model: BaseLLM,
        returnMap: true,
        useSpecs: true,
      });
@@ -273,6 +234,7 @@ describe('Tool Handlers', () => {
      process.env.SD_WEBUI_URL = mockCredential;
      toolFunctions = await loadTools({
        user: fakeUser._id,
+        model: BaseLLM,
        tools: ['stable-diffusion'],
        functions: true,
        returnMap: true,
--- a/api/cache/banViolation.js
+++ b/api/cache/banViolation.js
@@ -1,9 +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 { logger } = require('~/config');

 const { BAN_VIOLATIONS, BAN_INTERVAL } = process.env ?? {};
 const interval = math(BAN_INTERVAL, 20);
@@ -33,6 +32,7 @@ const banViolation = async (req, res, errorMessage) => {
  if (!isEnabled(BAN_VIOLATIONS)) {
    return;
  }
+
  if (!errorMessage) {
    return;
  }
@@ -51,6 +51,7 @@ const banViolation = async (req, res, errorMessage) => {

  const banLogs = getLogStores(ViolationTypes.BAN);
  const duration = errorMessage.duration || banLogs.opts.ttl;
+
  if (duration <= 0) {
    return;
  }
--- a/api/cache/banViolation.spec.js
+++ b/api/cache/banViolation.spec.js
@@ -1,28 +1,48 @@
-const mongoose = require('mongoose');
-const { MongoMemoryServer } = require('mongodb-memory-server');
 const banViolation = require('./banViolation');

-// Mock deleteAllUserSessions since we're testing ban logic, not session deletion
-jest.mock('~/models', () => ({
-  ...jest.requireActual('~/models'),
-  deleteAllUserSessions: jest.fn().mockResolvedValue(true),
-}));
+jest.mock('keyv');
+jest.mock('../models/Session');
+// Mocking the getLogStores function
+jest.mock('./getLogStores', () => {
+  return jest.fn().mockImplementation(() => {
+    const EventEmitter = require('events');
+    const { CacheKeys } = require('librechat-data-provider');
+    const math = require('../server/utils/math');
+    const mockGet = jest.fn();
+    const mockSet = jest.fn();
+    class KeyvMongo extends EventEmitter {
+      constructor(url = 'mongodb://127.0.0.1:27017', options) {
+        super();
+        this.ttlSupport = false;
+        url = url ?? {};
+        if (typeof url === 'string') {
+          url = { url };
+        }
+        if (url.uri) {
+          url = { url: url.uri, ...url };
+        }
+        this.opts = {
+          url,
+          collection: 'keyv',
+          ...url,
+          ...options,
+        };
+      }
+
+      get = mockGet;
+      set = mockSet;
+    }
+
+    return new KeyvMongo('', {
+      namespace: CacheKeys.BANS,
+      ttl: math(process.env.BAN_DURATION, 7200000),
+    });
+  });
+});

 describe('banViolation', () => {
-  let mongoServer;
  let req, res, errorMessage;

-  beforeAll(async () => {
-    mongoServer = await MongoMemoryServer.create();
-    const mongoUri = mongoServer.getUri();
-    await mongoose.connect(mongoUri);
-  });
-
-  afterAll(async () => {
-    await mongoose.disconnect();
-    await mongoServer.stop();
-  });
-
  beforeEach(() => {
    req = {
      ip: '127.0.0.1',
@@ -35,7 +55,7 @@ describe('banViolation', () => {
    };
    errorMessage = {
      type: 'someViolation',
-      user_id: new mongoose.Types.ObjectId().toString(), // Use valid ObjectId
+      user_id: '12345',
      prev_count: 0,
      violation_count: 0,
    };
--- a/api/cache/clearPendingReq.js
+++ b/api/cache/clearPendingReq.js
@@ -1,5 +1,5 @@
-const { isEnabled } = require('@librechat/api');
 const { Time, CacheKeys } = require('librechat-data-provider');
+const { isEnabled } = require('~/server/utils');
 const getLogStores = require('./getLogStores');

 const { USE_REDIS, LIMIT_CONCURRENT_MESSAGES } = process.env ?? {};
--- a/api/cache/getLogStores.js
+++ b/api/cache/getLogStores.js
@@ -1,56 +1,108 @@
 const { Keyv } = require('keyv');
-const { Time, CacheKeys, ViolationTypes } = require('librechat-data-provider');
-const {
-  logFile,
-  keyvMongo,
-  cacheConfig,
-  sessionCache,
-  standardCache,
-  violationCache,
-} = require('@librechat/api');
+const { CacheKeys, ViolationTypes, Time } = require('librechat-data-provider');
+const { logFile, violationFile } = require('./keyvFiles');
+const { math, isEnabled } = require('~/server/utils');
+const keyvRedis = require('./keyvRedis');
+const keyvMongo = require('./keyvMongo');
+
+const { BAN_DURATION, USE_REDIS, DEBUG_MEMORY_CACHE, CI } = process.env ?? {};
+
+const duration = math(BAN_DURATION, 7200000);
+const isRedisEnabled = isEnabled(USE_REDIS);
+const debugMemoryCache = isEnabled(DEBUG_MEMORY_CACHE);
+
+const createViolationInstance = (namespace) => {
+  const config = isRedisEnabled ? { store: keyvRedis } : { store: violationFile, namespace };
+  return new Keyv(config);
+};
+
+// Serve cache from memory so no need to clear it on startup/exit
+const pending_req = isRedisEnabled
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: CacheKeys.PENDING_REQ });
+
+const config = isRedisEnabled
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: CacheKeys.CONFIG_STORE });
+
+const roles = isRedisEnabled
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: CacheKeys.ROLES });
+
+const audioRuns = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.TEN_MINUTES })
+  : new Keyv({ namespace: CacheKeys.AUDIO_RUNS, ttl: Time.TEN_MINUTES });
+
+const messages = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.ONE_MINUTE })
+  : new Keyv({ namespace: CacheKeys.MESSAGES, ttl: Time.ONE_MINUTE });
+
+const flows = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.TWO_MINUTES })
+  : new Keyv({ namespace: CacheKeys.FLOWS, ttl: Time.ONE_MINUTE * 3 });
+
+const tokenConfig = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.THIRTY_MINUTES })
+  : new Keyv({ namespace: CacheKeys.TOKEN_CONFIG, ttl: Time.THIRTY_MINUTES });
+
+const genTitle = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.TWO_MINUTES })
+  : new Keyv({ namespace: CacheKeys.GEN_TITLE, ttl: Time.TWO_MINUTES });
+
+const s3ExpiryInterval = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.THIRTY_MINUTES })
+  : new Keyv({ namespace: CacheKeys.S3_EXPIRY_INTERVAL, ttl: Time.THIRTY_MINUTES });
+
+const modelQueries = isEnabled(process.env.USE_REDIS)
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: CacheKeys.MODEL_QUERIES });
+
+const abortKeys = isRedisEnabled
+  ? new Keyv({ store: keyvRedis })
+  : new Keyv({ namespace: CacheKeys.ABORT_KEYS, ttl: Time.TEN_MINUTES });
+
+const openIdExchangedTokensCache = isRedisEnabled
+  ? new Keyv({ store: keyvRedis, ttl: Time.TEN_MINUTES })
+  : new Keyv({ namespace: CacheKeys.OPENID_EXCHANGED_TOKENS, ttl: Time.TEN_MINUTES });

 const namespaces = {
-  [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({
+  [CacheKeys.ROLES]: roles,
+  [CacheKeys.CONFIG_STORE]: config,
+  [CacheKeys.PENDING_REQ]: pending_req,
+  [ViolationTypes.BAN]: new Keyv({ store: keyvMongo, namespace: CacheKeys.BANS, ttl: duration }),
+  [CacheKeys.ENCODED_DOMAINS]: new Keyv({
    store: keyvMongo,
-    namespace: CacheKeys.BANS,
-    ttl: cacheConfig.BAN_DURATION,
+    namespace: CacheKeys.ENCODED_DOMAINS,
+    ttl: 0,
  }),
-
-  [CacheKeys.OPENID_SESSION]: sessionCache(CacheKeys.OPENID_SESSION),
-  [CacheKeys.SAML_SESSION]: sessionCache(CacheKeys.SAML_SESSION),
-
-  [CacheKeys.ROLES]: standardCache(CacheKeys.ROLES),
-  [CacheKeys.APP_CONFIG]: standardCache(CacheKeys.APP_CONFIG),
-  [CacheKeys.CONFIG_STORE]: standardCache(CacheKeys.CONFIG_STORE),
-  [CacheKeys.PENDING_REQ]: standardCache(CacheKeys.PENDING_REQ),
-  [CacheKeys.ENCODED_DOMAINS]: new Keyv({ store: keyvMongo, namespace: CacheKeys.ENCODED_DOMAINS }),
-  [CacheKeys.ABORT_KEYS]: standardCache(CacheKeys.ABORT_KEYS, Time.TEN_MINUTES),
-  [CacheKeys.TOKEN_CONFIG]: standardCache(CacheKeys.TOKEN_CONFIG, Time.THIRTY_MINUTES),
-  [CacheKeys.GEN_TITLE]: standardCache(CacheKeys.GEN_TITLE, Time.TWO_MINUTES),
-  [CacheKeys.S3_EXPIRY_INTERVAL]: standardCache(CacheKeys.S3_EXPIRY_INTERVAL, Time.THIRTY_MINUTES),
-  [CacheKeys.MODEL_QUERIES]: standardCache(CacheKeys.MODEL_QUERIES),
-  [CacheKeys.AUDIO_RUNS]: standardCache(CacheKeys.AUDIO_RUNS, Time.TEN_MINUTES),
-  [CacheKeys.MESSAGES]: standardCache(CacheKeys.MESSAGES, Time.ONE_MINUTE),
-  [CacheKeys.FLOWS]: standardCache(CacheKeys.FLOWS, Time.ONE_MINUTE * 3),
-  [CacheKeys.OPENID_EXCHANGED_TOKENS]: standardCache(
-    CacheKeys.OPENID_EXCHANGED_TOKENS,
-    Time.TEN_MINUTES,
+  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,
  ),
+  [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,
 };

 /**
@@ -59,10 +111,7 @@ const namespaces = {
 */
 function getTTLStores() {
  return Object.values(namespaces).filter(
-    (store) =>
-      store instanceof Keyv &&
-      parseInt(store.opts?.ttl ?? '0') > 0 &&
-      !store.opts?.store?.constructor?.name?.includes('Redis'), // Only include non-Redis stores
+    (store) => store instanceof Keyv && typeof store.opts?.ttl === 'number' && store.opts.ttl > 0,
  );
 }

@@ -98,18 +147,18 @@ async function clearExpiredFromCache(cache) {
      if (data?.expires && data.expires <= expiryTime) {
        const deleted = await cache.opts.store.delete(key);
        if (!deleted) {
-          cacheConfig.DEBUG_MEMORY_CACHE &&
+          debugMemoryCache &&
            console.warn(`[Cache] Error deleting entry: ${key} from ${cache.opts.namespace}`);
          continue;
        }
        cleared++;
      }
    } catch (error) {
-      cacheConfig.DEBUG_MEMORY_CACHE &&
+      debugMemoryCache &&
        console.log(`[Cache] Error processing entry from ${cache.opts.namespace}:`, error);
      const deleted = await cache.opts.store.delete(key);
      if (!deleted) {
-        cacheConfig.DEBUG_MEMORY_CACHE &&
+        debugMemoryCache &&
          console.warn(`[Cache] Error deleting entry: ${key} from ${cache.opts.namespace}`);
        continue;
      }
@@ -118,7 +167,7 @@ async function clearExpiredFromCache(cache) {
  }

  if (cleared > 0) {
-    cacheConfig.DEBUG_MEMORY_CACHE &&
+    debugMemoryCache &&
      console.log(
        `[Cache] Cleared ${cleared} entries older than ${ttl}ms from ${cache.opts.namespace}`,
      );
@@ -159,7 +208,7 @@ async function clearAllExpiredFromCache() {
  }
 }

-if (!cacheConfig.USE_REDIS && !cacheConfig.CI) {
+if (!isRedisEnabled && !isEnabled(CI)) {
  /** @type {Set<NodeJS.Timeout>} */
  const cleanupIntervals = new Set();

@@ -170,7 +219,7 @@ if (!cacheConfig.USE_REDIS && !cacheConfig.CI) {

  cleanupIntervals.add(cleanup);

-  if (cacheConfig.DEBUG_MEMORY_CACHE) {
+  if (debugMemoryCache) {
    const monitor = setInterval(() => {
      const ttlStores = getTTLStores();
      const memory = process.memoryUsage();
@@ -191,13 +240,13 @@ if (!cacheConfig.USE_REDIS && !cacheConfig.CI) {
  }

  const dispose = () => {
-    cacheConfig.DEBUG_MEMORY_CACHE && console.log('[Cache] Cleaning up and shutting down...');
+    debugMemoryCache && console.log('[Cache] Cleaning up and shutting down...');
    cleanupIntervals.forEach((interval) => clearInterval(interval));
    cleanupIntervals.clear();

    // One final cleanup before exit
    clearAllExpiredFromCache().then(() => {
-      cacheConfig.DEBUG_MEMORY_CACHE && console.log('[Cache] Final cleanup completed');
+      debugMemoryCache && console.log('[Cache] Final cleanup completed');
      process.exit(0);
    });
  };
--- a/api/cache/index.js
+++ b/api/cache/index.js
@@ -1,4 +1,5 @@
+const keyvFiles = require('./keyvFiles');
 const getLogStores = require('./getLogStores');
 const logViolation = require('./logViolation');

-module.exports = { getLogStores, logViolation };
+module.exports = { ...keyvFiles, getLogStores, logViolation };
--- a/api/cache/ioredisClient.js
+++ b/api/cache/ioredisClient.js
@@ -0,0 +1,92 @@
+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;
--- a/api/cache/keyvFiles.js
+++ b/api/cache/keyvFiles.js
@@ -0,0 +1,9 @@
+const { KeyvFile } = require('keyv-file');
+
+const logFile = new KeyvFile({ filename: './data/logs.json' }).setMaxListeners(20);
+const violationFile = new KeyvFile({ filename: './data/violations.json' }).setMaxListeners(20);
+
+module.exports = {
+  logFile,
+  violationFile,
+};
--- a/packages/api/src/cache/keyvMongo.ts
+++ b/packages/api/src/cache/keyvMongo.ts
@@ -1,69 +1,65 @@
-import mongoose from 'mongoose';
-import { EventEmitter } from 'events';
-import { GridFSBucket } from 'mongodb';
-import { logger } from '@librechat/data-schemas';
-import type { Db, ReadPreference, Collection } from 'mongodb';
+// api/cache/keyvMongo.js
+const mongoose = require('mongoose');
+const EventEmitter = require('events');
+const { GridFSBucket } = require('mongodb');
+const { logger } = require('~/config');

-interface KeyvMongoOptions {
-  url?: string;
-  collection?: string;
-  useGridFS?: boolean;
-  readPreference?: ReadPreference;
-}
-
-interface GridFSClient {
-  bucket: GridFSBucket;
-  store: Collection;
-  db: Db;
-}
-
-interface CollectionClient {
-  store: Collection;
-  db: Db;
-}
-
-type Client = GridFSClient | CollectionClient;
-
-const storeMap = new Map<string, Client>();
+const storeMap = new Map();

 class KeyvMongoCustom extends EventEmitter {
-  private opts: KeyvMongoOptions;
-  public ttlSupport: boolean;
-  public namespace?: string;
-
-  constructor(options: KeyvMongoOptions = {}) {
+  constructor(url, options = {}) {
    super();

+    url = url || {};
+    if (typeof url === 'string') {
+      url = { url };
+    }
+    if (url.uri) {
+      url = { url: url.uri, ...url };
+    }
+
    this.opts = {
      url: 'mongodb://127.0.0.1:27017',
      collection: 'keyv',
+      ...url,
      ...options,
    };

    this.ttlSupport = false;
+
+    // Filter valid options
+    const keyvMongoKeys = new Set([
+      'url',
+      'collection',
+      'namespace',
+      'serialize',
+      'deserialize',
+      'uri',
+      'useGridFS',
+      'dialect',
+    ]);
+    this.opts = Object.fromEntries(Object.entries(this.opts).filter(([k]) => keyvMongoKeys.has(k)));
  }

  // Helper to access the store WITHOUT storing a promise on the instance
-  private async _getClient(): Promise<Client> {
+  _getClient() {
    const storeKey = `${this.opts.collection}:${this.opts.useGridFS ? 'gridfs' : 'collection'}`;

    // If we already have the store initialized, return it directly
    if (storeMap.has(storeKey)) {
-      return storeMap.get(storeKey)!;
+      return Promise.resolve(storeMap.get(storeKey));
    }

    // Check mongoose connection state
    if (mongoose.connection.readyState !== 1) {
-      throw new Error('Mongoose connection not ready. Ensure connectDb() is called first.');
+      return Promise.reject(
+        new Error('Mongoose connection not ready. Ensure connectDb() is called first.'),
+      );
    }

    try {
-      const db = mongoose.connection.db as unknown as Db | undefined;
-      if (!db) {
-        throw new Error('MongoDB database not available');
-      }
-
-      let client: Client;
+      const db = mongoose.connection.db;
+      let client;

      if (this.opts.useGridFS) {
        const bucket = new GridFSBucket(db, {
@@ -79,17 +75,17 @@ class KeyvMongoCustom extends EventEmitter {
      }

      storeMap.set(storeKey, client);
-      return client;
+      return Promise.resolve(client);
    } catch (error) {
      this.emit('error', error);
-      throw error;
+      return Promise.reject(error);
    }
  }

-  async get(key: string): Promise<unknown> {
+  async get(key) {
    const client = await this._getClient();

-    if (this.opts.useGridFS && this.isGridFSClient(client)) {
+    if (this.opts.useGridFS) {
      await client.store.updateOne(
        {
          filename: key,
@@ -104,7 +100,7 @@ class KeyvMongoCustom extends EventEmitter {
      const stream = client.bucket.openDownloadStreamByName(key);

      return new Promise((resolve) => {
-        const resp: Uint8Array[] = [];
+        const resp = [];
        stream.on('error', () => {
          resolve(undefined);
        });
@@ -114,7 +110,7 @@ class KeyvMongoCustom extends EventEmitter {
          resolve(data);
        });

-        stream.on('data', (chunk: Uint8Array) => {
+        stream.on('data', (chunk) => {
          resp.push(chunk);
        });
      });
@@ -129,7 +125,7 @@ class KeyvMongoCustom extends EventEmitter {
    return document.value;
  }

-  async getMany(keys: string[]): Promise<unknown[]> {
+  async getMany(keys) {
    const client = await this._getClient();

    if (this.opts.useGridFS) {
@@ -139,9 +135,9 @@ class KeyvMongoCustom extends EventEmitter {
      }

      const values = await Promise.allSettled(promises);
-      const data: unknown[] = [];
+      const data = [];
      for (const value of values) {
-        data.push(value.status === 'fulfilled' ? value.value : undefined);
+        data.push(value.value);
      }

      return data;
@@ -152,7 +148,7 @@ class KeyvMongoCustom extends EventEmitter {
      .project({ _id: 0, value: 1, key: 1 })
      .toArray();

-    const results: unknown[] = [...keys];
+    const results = [...keys];
    let i = 0;
    for (const key of keys) {
      const rowIndex = values.findIndex((row) => row.key === key);
@@ -163,11 +159,11 @@ class KeyvMongoCustom extends EventEmitter {
    return results;
  }

-  async set(key: string, value: string, ttl?: number): Promise<unknown> {
+  async set(key, value, ttl) {
    const client = await this._getClient();
    const expiresAt = typeof ttl === 'number' ? new Date(Date.now() + ttl) : null;

-    if (this.opts.useGridFS && this.isGridFSClient(client)) {
+    if (this.opts.useGridFS) {
      const stream = client.bucket.openUploadStream(key, {
        metadata: {
          expiresAt,
@@ -190,18 +186,20 @@ class KeyvMongoCustom extends EventEmitter {
    );
  }

-  async delete(key: string): Promise<boolean> {
+  async delete(key) {
+    if (typeof key !== 'string') {
+      return false;
+    }
+
    const client = await this._getClient();

-    if (this.opts.useGridFS && this.isGridFSClient(client)) {
+    if (this.opts.useGridFS) {
      try {
        const bucket = new GridFSBucket(client.db, {
          bucketName: this.opts.collection,
        });
        const files = await bucket.find({ filename: key }).toArray();
-        if (files.length > 0) {
-          await client.bucket.delete(files[0]._id);
-        }
+        await client.bucket.delete(files[0]._id);
        return true;
      } catch {
        return false;
@@ -212,10 +210,10 @@ class KeyvMongoCustom extends EventEmitter {
    return object.deletedCount > 0;
  }

-  async deleteMany(keys: string[]): Promise<boolean> {
+  async deleteMany(keys) {
    const client = await this._getClient();

-    if (this.opts.useGridFS && this.isGridFSClient(client)) {
+    if (this.opts.useGridFS) {
      const bucket = new GridFSBucket(client.db, {
        bucketName: this.opts.collection,
      });
@@ -232,17 +230,15 @@ class KeyvMongoCustom extends EventEmitter {
    return object.deletedCount > 0;
  }

-  async clear(): Promise<void> {
+  async clear() {
    const client = await this._getClient();

-    if (this.opts.useGridFS && this.isGridFSClient(client)) {
+    if (this.opts.useGridFS) {
      try {
        await client.bucket.drop();
-      } catch (error: unknown) {
+      } catch (error) {
        // Throw error if not "namespace not found" error
-        const errorCode =
-          error instanceof Error && 'code' in error ? (error as { code?: number }).code : undefined;
-        if (errorCode !== 26) {
+        if (!(error.code === 26)) {
          throw error;
        }
      }
@@ -253,7 +249,7 @@ class KeyvMongoCustom extends EventEmitter {
    });
  }

-  async has(key: string): Promise<boolean> {
+  async has(key) {
    const client = await this._getClient();
    const filter = { [this.opts.useGridFS ? 'filename' : 'key']: { $eq: key } };
    const document = await client.store.countDocuments(filter, { limit: 1 });
@@ -261,14 +257,10 @@ class KeyvMongoCustom extends EventEmitter {
  }

  // No-op disconnect
-  async disconnect(): Promise<boolean> {
+  async disconnect() {
    // This is a no-op since we don't want to close the shared mongoose connection
    return true;
  }
-
-  private isGridFSClient(client: Client): client is GridFSClient {
-    return (client as GridFSClient).bucket != null;
-  }
 }

 const keyvMongo = new KeyvMongoCustom({
@@ -277,4 +269,4 @@ const keyvMongo = new KeyvMongoCustom({

 keyvMongo.on('error', (err) => logger.error('KeyvMongo connection error:', err));

-export default keyvMongo;
+module.exports = keyvMongo;
--- a/api/cache/keyvRedis.js
+++ b/api/cache/keyvRedis.js
@@ -0,0 +1,109 @@
+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;
--- a/api/cache/logViolation.js
+++ b/api/cache/logViolation.js
@@ -1,5 +1,4 @@
-const { isEnabled } = require('@librechat/api');
-const { ViolationTypes } = require('librechat-data-provider');
+const { isEnabled } = require('~/server/utils');
 const getLogStores = require('./getLogStores');
 const banViolation = require('./banViolation');

@@ -10,14 +9,14 @@ 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 | string} [score=1] - The severity of the violation. Defaults to 1
+ * @param {number} [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(ViolationTypes.GENERAL);
+  const logs = getLogStores('general');
  const violationLogs = getLogStores(type);
  const key = isEnabled(process.env.USE_REDIS) ? `${type}:${userId}` : userId;

--- a/api/config/index.js
+++ b/api/config/index.js
@@ -1,13 +1,28 @@
+const axios = require('axios');
 const { EventSource } = require('eventsource');
-const { Time } = require('librechat-data-provider');
-const { MCPManager, FlowStateManager, OAuthReconnectionManager } = require('@librechat/api');
+const { Time, CacheKeys } = require('librechat-data-provider');
+const { MCPManager, FlowStateManager } = require('librechat-mcp');
 const logger = require('./winston');

 global.EventSource = EventSource;

 /** @type {MCPManager} */
+let mcpManager = null;
 let flowManager = null;

+/**
+ * @param {string} [userId] - Optional user ID, to avoid disconnecting the current user.
+ * @returns {MCPManager}
+ */
+function getMCPManager(userId) {
+  if (!mcpManager) {
+    mcpManager = MCPManager.getInstance(logger);
+  } else {
+    mcpManager.checkIdleConnections(userId);
+  }
+  return mcpManager;
+}
+
 /**
 * @param {Keyv} flowsCache
 * @returns {FlowStateManager}
@@ -16,16 +31,66 @@ function getFlowStateManager(flowsCache) {
  if (!flowManager) {
    flowManager = new FlowStateManager(flowsCache, {
      ttl: Time.ONE_MINUTE * 3,
+      logger,
    });
  }
  return flowManager;
 }

+/**
+ * Sends message data in Server Sent Events format.
+ * @param {ServerResponse} res - The server response.
+ * @param {{ data: string | Record<string, unknown>, event?: string }} event - The message event.
+ * @param {string} event.event - The type of event.
+ * @param {string} event.data - The message to be sent.
+ */
+const sendEvent = (res, event) => {
+  if (typeof event.data === 'string' && event.data.length === 0) {
+    return;
+  }
+  res.write(`event: message\ndata: ${JSON.stringify(event)}\n\n`);
+};
+
+/**
+ * Creates and configures an Axios instance with optional proxy settings.
+ *
+ * @typedef {import('axios').AxiosInstance} AxiosInstance
+ * @typedef {import('axios').AxiosProxyConfig} AxiosProxyConfig
+ *
+ * @returns {AxiosInstance} A configured Axios instance
+ * @throws {Error} If there's an issue creating the Axios instance or parsing the proxy URL
+ */
+function createAxiosInstance() {
+  const instance = axios.create();
+
+  if (process.env.proxy) {
+    try {
+      const url = new URL(process.env.proxy);
+
+      /** @type {AxiosProxyConfig} */
+      const proxyConfig = {
+        host: url.hostname.replace(/^\[|\]$/g, ''),
+        protocol: url.protocol.replace(':', ''),
+      };
+
+      if (url.port) {
+        proxyConfig.port = parseInt(url.port, 10);
+      }
+
+      instance.defaults.proxy = proxyConfig;
+    } catch (error) {
+      console.error('Error parsing proxy URL:', error);
+      throw new Error(`Invalid proxy URL: ${process.env.proxy}`);
+    }
+  }
+
+  return instance;
+}
+
 module.exports = {
  logger,
-  createMCPManager: MCPManager.createInstance,
-  getMCPManager: MCPManager.getInstance,
+  sendEvent,
+  getMCPManager,
+  createAxiosInstance,
  getFlowStateManager,
-  createOAuthReconnectionManager: OAuthReconnectionManager.createInstance,
-  getOAuthReconnectionManager: OAuthReconnectionManager.getInstance,
 };
--- a/packages/api/src/utils/axios.spec.ts
+++ b/packages/api/src/utils/axios.spec.ts
@@ -1,6 +1,7 @@
-import axios from 'axios';
-import { createAxiosInstance } from './axios';
+const axios = require('axios');
+const { createAxiosInstance } = require('./index');

+// Mock axios
 jest.mock('axios', () => ({
  interceptors: {
    request: { use: jest.fn(), eject: jest.fn() },
@@ -19,13 +20,7 @@ jest.mock('axios', () => ({
  post: jest.fn().mockResolvedValue({ data: {} }),
  put: jest.fn().mockResolvedValue({ data: {} }),
  delete: jest.fn().mockResolvedValue({ data: {} }),
-  reset: jest.fn().mockImplementation(function (this: {
-    get: jest.Mock;
-    post: jest.Mock;
-    put: jest.Mock;
-    delete: jest.Mock;
-    create: jest.Mock;
-  }) {
+  reset: jest.fn().mockImplementation(function () {
    this.get.mockClear();
    this.post.mockClear();
    this.put.mockClear();
--- a/api/db/connect.js
+++ b/api/db/connect.js
@@ -1,79 +0,0 @@
-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
- * during API Route usage.
- */
-let cached = global.mongoose;
-
-if (!cached) {
-  cached = global.mongoose = { conn: null, promise: null };
-}
-
-async function connectDb() {
-  if (cached.conn && cached.conn?._readyState === 1) {
-    return cached.conn;
-  }
-
-  const disconnected = cached.conn && cached.conn?._readyState !== 1;
-  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;
-    });
-  }
-  cached.conn = await cached.promise;
-
-  return cached.conn;
-}
-
-module.exports = {
-  connectDb,
-};
--- a/api/db/index.js
+++ b/api/db/index.js
@@ -1,8 +0,0 @@
-const mongoose = require('mongoose');
-const { createModels } = require('@librechat/data-schemas');
-const { connectDb } = require('./connect');
-const indexSync = require('./indexSync');
-
-createModels(mongoose);
-
-module.exports = { connectDb, indexSync };
--- a/api/db/indexSync.js
+++ b/api/db/indexSync.js
@@ -1,360 +0,0 @@
-const mongoose = require('mongoose');
-const { MeiliSearch } = require('meilisearch');
-const { logger } = require('@librechat/data-schemas');
-const { CacheKeys } = require('librechat-data-provider');
-const { isEnabled, FlowStateManager } = require('@librechat/api');
-const { getLogStores } = require('~/cache');
-
-const Conversation = mongoose.models.Conversation;
-const Message = mongoose.models.Message;
-
-const searchEnabled = isEnabled(process.env.SEARCH);
-const indexingDisabled = isEnabled(process.env.MEILI_NO_SYNC);
-let currentTimeout = null;
-
-class MeiliSearchClient {
-  static instance = null;
-
-  static getInstance() {
-    if (!MeiliSearchClient.instance) {
-      if (!process.env.MEILI_HOST || !process.env.MEILI_MASTER_KEY) {
-        throw new Error('Meilisearch configuration is missing.');
-      }
-      MeiliSearchClient.instance = new MeiliSearch({
-        host: process.env.MEILI_HOST,
-        apiKey: process.env.MEILI_MASTER_KEY,
-      });
-    }
-    return MeiliSearchClient.instance;
-  }
-}
-
-/**
- * Deletes documents from MeiliSearch index that are missing the user field
- * @param {import('meilisearch').Index} index - MeiliSearch index instance
- * @param {string} indexName - Name of the index for logging
- * @returns {Promise<number>} - Number of documents deleted
- */
-async function deleteDocumentsWithoutUserField(index, indexName) {
-  let deletedCount = 0;
-  let offset = 0;
-  const batchSize = 1000;
-
-  try {
-    while (true) {
-      const searchResult = await index.search('', {
-        limit: batchSize,
-        offset: offset,
-      });
-
-      if (searchResult.hits.length === 0) {
-        break;
-      }
-
-      const idsToDelete = searchResult.hits.filter((hit) => !hit.user).map((hit) => hit.id);
-
-      if (idsToDelete.length > 0) {
-        logger.info(
-          `[indexSync] Deleting ${idsToDelete.length} documents without user field from ${indexName} index`,
-        );
-        await index.deleteDocuments(idsToDelete);
-        deletedCount += idsToDelete.length;
-      }
-
-      if (searchResult.hits.length < batchSize) {
-        break;
-      }
-
-      offset += batchSize;
-    }
-
-    if (deletedCount > 0) {
-      logger.info(`[indexSync] Deleted ${deletedCount} orphaned documents from ${indexName} index`);
-    }
-  } catch (error) {
-    logger.error(`[indexSync] Error deleting documents from ${indexName}:`, error);
-  }
-
-  return deletedCount;
-}
-
-/**
- * Ensures indexes have proper filterable attributes configured and checks if documents have user field
- * @param {MeiliSearch} client - MeiliSearch client instance
- * @returns {Promise<{settingsUpdated: boolean, orphanedDocsFound: boolean}>} - Status of what was done
- */
-async function ensureFilterableAttributes(client) {
-  let settingsUpdated = false;
-  let hasOrphanedDocs = false;
-
-  try {
-    // Check and update messages index
-    try {
-      const messagesIndex = client.index('messages');
-      const settings = await messagesIndex.getSettings();
-
-      if (!settings.filterableAttributes || !settings.filterableAttributes.includes('user')) {
-        logger.info('[indexSync] Configuring messages index to filter by user...');
-        await messagesIndex.updateSettings({
-          filterableAttributes: ['user'],
-        });
-        logger.info('[indexSync] Messages index configured for user filtering');
-        settingsUpdated = true;
-      }
-
-      // Check if existing documents have user field indexed
-      try {
-        const searchResult = await messagesIndex.search('', { limit: 1 });
-        if (searchResult.hits.length > 0 && !searchResult.hits[0].user) {
-          logger.info(
-            '[indexSync] Existing messages missing user field, will clean up orphaned documents...',
-          );
-          hasOrphanedDocs = true;
-        }
-      } catch (searchError) {
-        logger.debug('[indexSync] Could not check message documents:', searchError.message);
-      }
-    } catch (error) {
-      if (error.code !== 'index_not_found') {
-        logger.warn('[indexSync] Could not check/update messages index settings:', error.message);
-      }
-    }
-
-    // Check and update conversations index
-    try {
-      const convosIndex = client.index('convos');
-      const settings = await convosIndex.getSettings();
-
-      if (!settings.filterableAttributes || !settings.filterableAttributes.includes('user')) {
-        logger.info('[indexSync] Configuring convos index to filter by user...');
-        await convosIndex.updateSettings({
-          filterableAttributes: ['user'],
-        });
-        logger.info('[indexSync] Convos index configured for user filtering');
-        settingsUpdated = true;
-      }
-
-      // Check if existing documents have user field indexed
-      try {
-        const searchResult = await convosIndex.search('', { limit: 1 });
-        if (searchResult.hits.length > 0 && !searchResult.hits[0].user) {
-          logger.info(
-            '[indexSync] Existing conversations missing user field, will clean up orphaned documents...',
-          );
-          hasOrphanedDocs = true;
-        }
-      } catch (searchError) {
-        logger.debug('[indexSync] Could not check conversation documents:', searchError.message);
-      }
-    } catch (error) {
-      if (error.code !== 'index_not_found') {
-        logger.warn('[indexSync] Could not check/update convos index settings:', error.message);
-      }
-    }
-
-    // If either index has orphaned documents, clean them up (but don't force resync)
-    if (hasOrphanedDocs) {
-      try {
-        const messagesIndex = client.index('messages');
-        await deleteDocumentsWithoutUserField(messagesIndex, 'messages');
-      } catch (error) {
-        logger.debug('[indexSync] Could not clean up messages:', error.message);
-      }
-
-      try {
-        const convosIndex = client.index('convos');
-        await deleteDocumentsWithoutUserField(convosIndex, 'convos');
-      } catch (error) {
-        logger.debug('[indexSync] Could not clean up convos:', error.message);
-      }
-
-      logger.info('[indexSync] Orphaned documents cleaned up without forcing resync.');
-    }
-
-    if (settingsUpdated) {
-      logger.info('[indexSync] Index settings updated. Full re-sync will be triggered.');
-    }
-  } catch (error) {
-    logger.error('[indexSync] Error ensuring filterable attributes:', error);
-  }
-
-  return { settingsUpdated, orphanedDocsFound: hasOrphanedDocs };
-}
-
-/**
- * Performs the actual sync operations for messages and conversations
- * @param {FlowStateManager} flowManager - Flow state manager instance
- * @param {string} flowId - Flow identifier
- * @param {string} flowType - Flow type
- */
-async function performSync(flowManager, flowId, flowType) {
-  try {
-    const client = MeiliSearchClient.getInstance();
-
-    const { status } = await client.health();
-    if (status !== 'available') {
-      throw new Error('Meilisearch not available');
-    }
-
-    if (indexingDisabled === true) {
-      logger.info('[indexSync] Indexing is disabled, skipping...');
-      return { messagesSync: false, convosSync: false };
-    }
-
-    /** Ensures indexes have proper filterable attributes configured */
-    const { settingsUpdated, orphanedDocsFound: _orphanedDocsFound } =
-      await ensureFilterableAttributes(client);
-
-    let messagesSync = false;
-    let convosSync = false;
-
-    // Only reset flags if settings were actually updated (not just for orphaned doc cleanup)
-    if (settingsUpdated) {
-      logger.info(
-        '[indexSync] Settings updated. Forcing full re-sync to reindex with new configuration...',
-      );
-
-      // Reset sync flags to force full re-sync
-      await Message.collection.updateMany({ _meiliIndex: true }, { $set: { _meiliIndex: false } });
-      await Conversation.collection.updateMany(
-        { _meiliIndex: true },
-        { $set: { _meiliIndex: false } },
-      );
-    }
-
-    // Check if we need to sync messages
-    const messageProgress = await Message.getSyncProgress();
-    if (!messageProgress.isComplete || settingsUpdated) {
-      logger.info(
-        `[indexSync] Messages need syncing: ${messageProgress.totalProcessed}/${messageProgress.totalDocuments} indexed`,
-      );
-
-      // Check if we should do a full sync or incremental
-      const messageCount = await Message.countDocuments();
-      const messagesIndexed = messageProgress.totalProcessed;
-      const syncThreshold = parseInt(process.env.MEILI_SYNC_THRESHOLD || '1000', 10);
-
-      if (messageCount - messagesIndexed > syncThreshold) {
-        logger.info('[indexSync] Starting full message sync due to large difference');
-        await Message.syncWithMeili();
-        messagesSync = true;
-      } else if (messageCount !== messagesIndexed) {
-        logger.warn('[indexSync] Messages out of sync, performing incremental sync');
-        await Message.syncWithMeili();
-        messagesSync = true;
-      }
-    } else {
-      logger.info(
-        `[indexSync] Messages are fully synced: ${messageProgress.totalProcessed}/${messageProgress.totalDocuments}`,
-      );
-    }
-
-    // Check if we need to sync conversations
-    const convoProgress = await Conversation.getSyncProgress();
-    if (!convoProgress.isComplete || settingsUpdated) {
-      logger.info(
-        `[indexSync] Conversations need syncing: ${convoProgress.totalProcessed}/${convoProgress.totalDocuments} indexed`,
-      );
-
-      const convoCount = await Conversation.countDocuments();
-      const convosIndexed = convoProgress.totalProcessed;
-      const syncThreshold = parseInt(process.env.MEILI_SYNC_THRESHOLD || '1000', 10);
-
-      if (convoCount - convosIndexed > syncThreshold) {
-        logger.info('[indexSync] Starting full conversation sync due to large difference');
-        await Conversation.syncWithMeili();
-        convosSync = true;
-      } else if (convoCount !== convosIndexed) {
-        logger.warn('[indexSync] Convos out of sync, performing incremental sync');
-        await Conversation.syncWithMeili();
-        convosSync = true;
-      }
-    } else {
-      logger.info(
-        `[indexSync] Conversations are fully synced: ${convoProgress.totalProcessed}/${convoProgress.totalDocuments}`,
-      );
-    }
-
-    return { messagesSync, convosSync };
-  } finally {
-    if (indexingDisabled === true) {
-      logger.info('[indexSync] Indexing is disabled, skipping cleanup...');
-    } else if (flowManager && flowId && flowType) {
-      try {
-        await flowManager.deleteFlow(flowId, flowType);
-        logger.debug('[indexSync] Flow state cleaned up');
-      } catch (cleanupErr) {
-        logger.debug('[indexSync] Could not clean up flow state:', cleanupErr.message);
-      }
-    }
-  }
-}
-
-/**
- * Main index sync function that uses FlowStateManager to prevent concurrent execution
- */
-async function indexSync() {
-  if (!searchEnabled) {
-    return;
-  }
-
-  logger.info('[indexSync] Starting index synchronization check...');
-
-  // Get or create FlowStateManager instance
-  const flowsCache = getLogStores(CacheKeys.FLOWS);
-  if (!flowsCache) {
-    logger.warn('[indexSync] Flows cache not available, falling back to direct sync');
-    return await performSync(null, null, null);
-  }
-
-  const flowManager = new FlowStateManager(flowsCache, {
-    ttl: 60000 * 10, // 10 minutes TTL for sync operations
-  });
-
-  // Use a unique flow ID for the sync operation
-  const flowId = 'meili-index-sync';
-  const flowType = 'MEILI_SYNC';
-
-  try {
-    // This will only execute the handler if no other instance is running the sync
-    const result = await flowManager.createFlowWithHandler(flowId, flowType, () =>
-      performSync(flowManager, flowId, flowType),
-    );
-
-    if (result.messagesSync || result.convosSync) {
-      logger.info('[indexSync] Sync completed successfully');
-    } else {
-      logger.debug('[indexSync] No sync was needed');
-    }
-
-    return result;
-  } catch (err) {
-    if (err.message.includes('flow already exists')) {
-      logger.info('[indexSync] Sync already running on another instance');
-      return;
-    }
-
-    if (err.message.includes('not found')) {
-      logger.debug('[indexSync] Creating indices...');
-      currentTimeout = setTimeout(async () => {
-        try {
-          await Message.syncWithMeili();
-          await Conversation.syncWithMeili();
-        } catch (err) {
-          logger.error('[indexSync] Trouble creating indices, try restarting the server.', err);
-        }
-      }, 750);
-    } else if (err.message.includes('Meilisearch not configured')) {
-      logger.info('[indexSync] Meilisearch not configured, search will be disabled.');
-    } else {
-      logger.error('[indexSync] error', err);
-    }
-  }
-}
-
-process.on('exit', () => {
-  logger.debug('[indexSync] Clearing sync timeouts before exiting...');
-  clearTimeout(currentTimeout);
-});
-
-module.exports = indexSync;
--- a/api/db/models.js
+++ b/api/db/models.js
@@ -1,5 +0,0 @@
-const mongoose = require('mongoose');
-const { createModels } = require('@librechat/data-schemas');
-const models = createModels(mongoose);
-
-module.exports = { ...models };
--- a/api/jest.config.js
+++ b/api/jest.config.js
@@ -3,7 +3,6 @@ module.exports = {
  clearMocks: true,
  roots: ['<rootDir>'],
  coverageDirectory: 'coverage',
-  testTimeout: 30000, // 30 seconds timeout for all tests
  setupFiles: [
    './test/jestSetup.js',
    './test/__mocks__/logger.js',
--- a/api/lib/db/connectDb.js
+++ b/api/lib/db/connectDb.js
@@ -0,0 +1,45 @@
+require('dotenv').config();
+const mongoose = require('mongoose');
+const MONGO_URI = process.env.MONGO_URI;
+
+if (!MONGO_URI) {
+  throw new Error('Please define the MONGO_URI environment variable');
+}
+
+/**
+ * Global is used here to maintain a cached connection across hot reloads
+ * in development. This prevents connections growing exponentially
+ * during API Route usage.
+ */
+let cached = global.mongoose;
+
+if (!cached) {
+  cached = global.mongoose = { conn: null, promise: null };
+}
+
+async function connectDb() {
+  if (cached.conn && cached.conn?._readyState === 1) {
+    return cached.conn;
+  }
+
+  const disconnected = cached.conn && cached.conn?._readyState !== 1;
+  if (!cached.promise || disconnected) {
+    const opts = {
+      bufferCommands: false,
+      // useNewUrlParser: true,
+      // useUnifiedTopology: true,
+      // bufferMaxEntries: 0,
+      // useFindAndModify: true,
+      // useCreateIndex: true
+    };
+
+    mongoose.set('strictQuery', true);
+    cached.promise = mongoose.connect(MONGO_URI, opts).then((mongoose) => {
+      return mongoose;
+    });
+  }
+  cached.conn = await cached.promise;
+  return cached.conn;
+}
+
+module.exports = connectDb;
--- a/api/lib/db/index.js
+++ b/api/lib/db/index.js
@@ -0,0 +1,4 @@
+const connectDb = require('./connectDb');
+const indexSync = require('./indexSync');
+
+module.exports = { connectDb, indexSync };
--- a/api/lib/db/indexSync.js
+++ b/api/lib/db/indexSync.js
@@ -0,0 +1,89 @@
+const { MeiliSearch } = require('meilisearch');
+const { Conversation } = require('~/models/Conversation');
+const { Message } = require('~/models/Message');
+const { isEnabled } = require('~/server/utils');
+const { logger } = require('~/config');
+
+const searchEnabled = isEnabled(process.env.SEARCH);
+const indexingDisabled = isEnabled(process.env.MEILI_NO_SYNC);
+let currentTimeout = null;
+
+class MeiliSearchClient {
+  static instance = null;
+
+  static getInstance() {
+    if (!MeiliSearchClient.instance) {
+      if (!process.env.MEILI_HOST || !process.env.MEILI_MASTER_KEY) {
+        throw new Error('Meilisearch configuration is missing.');
+      }
+      MeiliSearchClient.instance = new MeiliSearch({
+        host: process.env.MEILI_HOST,
+        apiKey: process.env.MEILI_MASTER_KEY,
+      });
+    }
+    return MeiliSearchClient.instance;
+  }
+}
+
+async function indexSync() {
+  if (!searchEnabled) {
+    return;
+  }
+
+  try {
+    const client = MeiliSearchClient.getInstance();
+
+    const { status } = await client.health();
+    if (status !== 'available') {
+      throw new Error('Meilisearch not available');
+    }
+
+    if (indexingDisabled === true) {
+      logger.info('[indexSync] Indexing is disabled, skipping...');
+      return;
+    }
+
+    const messageCount = await Message.countDocuments();
+    const convoCount = await Conversation.countDocuments();
+    const messages = await client.index('messages').getStats();
+    const convos = await client.index('convos').getStats();
+    const messagesIndexed = messages.numberOfDocuments;
+    const convosIndexed = convos.numberOfDocuments;
+
+    logger.debug(`[indexSync] There are ${messageCount} messages and ${messagesIndexed} indexed`);
+    logger.debug(`[indexSync] There are ${convoCount} convos and ${convosIndexed} indexed`);
+
+    if (messageCount !== messagesIndexed) {
+      logger.debug('[indexSync] Messages out of sync, indexing');
+      Message.syncWithMeili();
+    }
+
+    if (convoCount !== convosIndexed) {
+      logger.debug('[indexSync] Convos out of sync, indexing');
+      Conversation.syncWithMeili();
+    }
+  } catch (err) {
+    if (err.message.includes('not found')) {
+      logger.debug('[indexSync] Creating indices...');
+      currentTimeout = setTimeout(async () => {
+        try {
+          await Message.syncWithMeili();
+          await Conversation.syncWithMeili();
+        } catch (err) {
+          logger.error('[indexSync] Trouble creating indices, try restarting the server.', err);
+        }
+      }, 750);
+    } else if (err.message.includes('Meilisearch not configured')) {
+      logger.info('[indexSync] Meilisearch not configured, search will be disabled.');
+    } else {
+      logger.error('[indexSync] error', err);
+    }
+  }
+}
+
+process.on('exit', () => {
+  logger.debug('[indexSync] Clearing sync timeouts before exiting...');
+  clearTimeout(currentTimeout);
+});
+
+module.exports = indexSync;
--- a/api/models/Action.js
+++ b/api/models/Action.js
@@ -1,4 +1,7 @@
-const { Action } = require('~/db/models');
+const mongoose = require('mongoose');
+const { actionSchema } = require('@librechat/data-schemas');
+
+const Action = mongoose.model('action', actionSchema);

 /**
 * Update an action with new data without overwriting existing properties,
--- a/api/models/Agent.js
+++ b/api/models/Agent.js
@@ -1,19 +1,21 @@
 const mongoose = require('mongoose');
 const crypto = require('node:crypto');
-const { logger } = require('@librechat/data-schemas');
-const { ResourceType, SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
-const { GLOBAL_PROJECT_NAME, EPHEMERAL_AGENT_ID, mcp_all, mcp_delimiter } =
+const { agentSchema } = require('@librechat/data-schemas');
+const { SystemRoles, Tools, actionDelimiter } = require('librechat-data-provider');
+const { GLOBAL_PROJECT_NAME, EPHEMERAL_AGENT_ID, mcp_delimiter } =
  require('librechat-data-provider').Constants;
+const { CONFIG_STORE, STARTUP_CONFIG } = require('librechat-data-provider').CacheKeys;
 const {
-  removeAgentFromAllProjects,
-  removeAgentIdsFromProject,
-  addAgentIdsToProject,
  getProjectByName,
+  addAgentIdsToProject,
+  removeAgentIdsFromProject,
+  removeAgentFromAllProjects,
 } = require('./Project');
-const { removeAllPermissions } = require('~/server/services/PermissionService');
-const { getMCPServerTools } = require('~/server/services/Config');
+const getLogStores = require('~/cache/getLogStores');
 const { getActions } = require('./Action');
-const { Agent } = require('~/db/models');
+const { logger } = require('~/config');
+
+const Agent = mongoose.model('agent', agentSchema);

 /**
 * Create an agent with the provided data.
@@ -22,7 +24,7 @@ const { Agent } = require('~/db/models');
 * @throws {Error} If the agent creation fails.
 */
 const createAgent = async (agentData) => {
-  const { author: _author, ...versionData } = agentData;
+  const { author, ...versionData } = agentData;
  const timestamp = new Date();
  const initialAgentData = {
    ...agentData,
@@ -33,9 +35,7 @@ const createAgent = async (agentData) => {
        updatedAt: timestamp,
      },
    ],
-    category: agentData.category || 'general',
  };
-
  return (await Agent.create(initialAgentData)).toObject();
 };

@@ -49,72 +49,46 @@ const createAgent = async (agentData) => {
 */
 const getAgent = async (searchParameter) => await Agent.findOne(searchParameter).lean();

-/**
- * Get multiple agent documents based on the provided search parameters.
- *
- * @param {Object} searchParameter - The search parameters to find agents.
- * @returns {Promise<Agent[]>} Array of agent documents as plain objects.
- */
-const getAgents = async (searchParameter) => await Agent.find(searchParameter).lean();
-
 /**
 * Load an agent based on the provided ID
 *
 * @param {Object} params
 * @param {ServerRequest} params.req
- * @param {string} params.spec
 * @param {string} params.agent_id
 * @param {string} params.endpoint
 * @param {import('@librechat/agents').ClientOptions} [params.model_parameters]
- * @returns {Promise<Agent|null>} The agent document as a plain object, or null if not found.
+ * @returns {Agent|null} The agent document as a plain object, or null if not found.
 */
-const loadEphemeralAgent = async ({ req, spec, agent_id, endpoint, model_parameters: _m }) => {
+const loadEphemeralAgent = ({ req, agent_id, endpoint, model_parameters: _m }) => {
  const { model, ...model_parameters } = _m;
-  const modelSpecs = req.config?.modelSpecs?.list;
-  /** @type {TModelSpec | null} */
-  let modelSpec = null;
-  if (spec != null && spec !== '') {
-    modelSpec = modelSpecs?.find((s) => s.name === spec) || null;
-  }
+  /** @type {Record<string, FunctionTool>} */
+  const availableTools = req.app.locals.availableTools;
  /** @type {TEphemeralAgent | null} */
  const ephemeralAgent = req.body.ephemeralAgent;
  const mcpServers = new Set(ephemeralAgent?.mcp);
-  if (modelSpec?.mcpServers) {
-    for (const mcpServer of modelSpec.mcpServers) {
-      mcpServers.add(mcpServer);
-    }
-  }
  /** @type {string[]} */
  const tools = [];
-  if (ephemeralAgent?.execute_code === true || modelSpec?.executeCode === true) {
+  if (ephemeralAgent?.execute_code === true) {
    tools.push(Tools.execute_code);
  }
-  if (ephemeralAgent?.file_search === true || modelSpec?.fileSearch === true) {
-    tools.push(Tools.file_search);
-  }
-  if (ephemeralAgent?.web_search === true || modelSpec?.webSearch === true) {
+  if (ephemeralAgent?.web_search === true) {
    tools.push(Tools.web_search);
  }

-  const addedServers = new Set();
  if (mcpServers.size > 0) {
-    for (const mcpServer of mcpServers) {
-      if (addedServers.has(mcpServer)) {
+    for (const toolName of Object.keys(availableTools)) {
+      if (!toolName.includes(mcp_delimiter)) {
        continue;
      }
-      const serverTools = await getMCPServerTools(mcpServer);
-      if (!serverTools) {
-        tools.push(`${mcp_all}${mcp_delimiter}${mcpServer}`);
-        addedServers.add(mcpServer);
-        continue;
+      const mcpServer = toolName.split(mcp_delimiter)?.[1];
+      if (mcpServer && mcpServers.has(mcpServer)) {
+        tools.push(toolName);
      }
-      tools.push(...Object.keys(serverTools));
-      addedServers.add(mcpServer);
    }
  }

  const instructions = req.body.promptPrefix;
-  const result = {
+  return {
    id: agent_id,
    instructions,
    provider: endpoint,
@@ -122,11 +96,6 @@ const loadEphemeralAgent = async ({ req, spec, agent_id, endpoint, model_paramet
    model,
    tools,
  };
-
-  if (ephemeralAgent?.artifacts != null && ephemeralAgent.artifacts) {
-    result.artifacts = ephemeralAgent.artifacts;
-  }
-  return result;
 };

 /**
@@ -134,18 +103,17 @@ const loadEphemeralAgent = async ({ req, spec, agent_id, endpoint, model_paramet
 *
 * @param {Object} params
 * @param {ServerRequest} params.req
- * @param {string} params.spec
 * @param {string} params.agent_id
 * @param {string} params.endpoint
 * @param {import('@librechat/agents').ClientOptions} [params.model_parameters]
 * @returns {Promise<Agent|null>} The agent document as a plain object, or null if not found.
 */
-const loadAgent = async ({ req, spec, agent_id, endpoint, model_parameters }) => {
+const loadAgent = async ({ req, agent_id, endpoint, model_parameters }) => {
  if (!agent_id) {
    return null;
  }
  if (agent_id === EPHEMERAL_AGENT_ID) {
-    return await loadEphemeralAgent({ req, spec, agent_id, endpoint, model_parameters });
+    return loadEphemeralAgent({ req, agent_id, endpoint, model_parameters });
  }
  const agent = await getAgent({
    id: agent_id,
@@ -156,7 +124,29 @@ const loadAgent = async ({ req, spec, agent_id, endpoint, model_parameters }) =>
  }

  agent.version = agent.versions ? agent.versions.length : 0;
-  return agent;
+
+  if (agent.author.toString() === req.user.id) {
+    return agent;
+  }
+
+  if (!agent.projectIds) {
+    return null;
+  }
+
+  const cache = getLogStores(CONFIG_STORE);
+  /** @type {TStartupConfig} */
+  const cachedStartupConfig = await cache.get(STARTUP_CONFIG);
+  let { instanceProjectId } = cachedStartupConfig ?? {};
+  if (!instanceProjectId) {
+    instanceProjectId = (await getProjectByName(GLOBAL_PROJECT_NAME, '_id'))._id.toString();
+  }
+
+  for (const projectObjectId of agent.projectIds) {
+    const projectId = projectObjectId.toString();
+    if (projectId === instanceProjectId) {
+      return agent;
+    }
+  }
 };

 /**
@@ -182,11 +172,12 @@ const isDuplicateVersion = (updateData, currentData, versions, actionsHash = nul
    'created_at',
    'updated_at',
    '__v',
+    'agent_ids',
    'versions',
    'actionsHash', // Exclude actionsHash from direct comparison
  ];

-  const { $push: _$push, $pull: _$pull, $addToSet: _$addToSet, ...directUpdates } = updateData;
+  const { $push, $pull, $addToSet, ...directUpdates } = updateData;

  if (Object.keys(directUpdates).length === 0 && !actionsHash) {
    return null;
@@ -205,116 +196,54 @@ const isDuplicateVersion = (updateData, currentData, versions, actionsHash = nul

  let isMatch = true;
  for (const field of importantFields) {
-    const wouldBeValue = wouldBeVersion[field];
-    const lastVersionValue = lastVersion[field];
-
-    // Skip if both are undefined/null
-    if (!wouldBeValue && !lastVersionValue) {
+    if (!wouldBeVersion[field] && !lastVersion[field]) {
      continue;
    }

-    // 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) {
+    if (Array.isArray(wouldBeVersion[field]) && Array.isArray(lastVersion[field])) {
+      if (wouldBeVersion[field].length !== lastVersion[field].length) {
        isMatch = false;
        break;
      }

      // Special handling for projectIds (MongoDB ObjectIds)
      if (field === 'projectIds') {
-        const wouldBeIds = wouldBeArr.map((id) => id.toString()).sort();
-        const versionIds = lastVersionArr.map((id) => id.toString()).sort();
+        const wouldBeIds = wouldBeVersion[field].map((id) => id.toString()).sort();
+        const versionIds = lastVersion[field].map((id) => id.toString()).sort();

        if (!wouldBeIds.every((id, i) => id === versionIds[i])) {
          isMatch = false;
          break;
        }
      }
-      // 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();
+      // 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();

        if (!sortedWouldBe.every((item, i) => item === sortedVersion[i])) {
          isMatch = false;
          break;
        }
      } else {
-        const sortedWouldBe = [...wouldBeArr].sort();
-        const sortedVersion = [...lastVersionArr].sort();
+        const sortedWouldBe = [...wouldBeVersion[field]].sort();
+        const sortedVersion = [...lastVersion[field]].sort();

        if (!sortedWouldBe.every((item, i) => item === sortedVersion[i])) {
          isMatch = false;
          break;
        }
      }
-    }
-    // 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;
-        }
+    } else if (field === 'model_parameters') {
+      const wouldBeParams = wouldBeVersion[field] || {};
+      const lastVersionParams = lastVersion[field] || {};
+      if (JSON.stringify(wouldBeParams) !== JSON.stringify(lastVersionParams)) {
        isMatch = false;
        break;
      }
+    } else if (wouldBeVersion[field] !== lastVersion[field]) {
+      isMatch = false;
+      break;
    }
  }

@@ -333,24 +262,16 @@ const isDuplicateVersion = (updateData, currentData, versions, actionsHash = nul
 * @param {Object} [options] - Optional configuration object.
 * @param {string} [options.updatingUserId] - The ID of the user performing the update (used for tracking non-author updates).
 * @param {boolean} [options.forceVersion] - Force creation of a new version even if no fields changed.
- * @param {boolean} [options.skipVersioning] - Skip version creation entirely (useful for isolated operations like sharing).
 * @returns {Promise<Agent>} The updated or newly created agent document as a plain object.
 * @throws {Error} If the update would create a duplicate version
 */
 const updateAgent = async (searchParameter, updateData, options = {}) => {
-  const { updatingUserId = null, forceVersion = false, skipVersioning = false } = options;
+  const { updatingUserId = null, forceVersion = false } = options;
  const mongoOptions = { new: true, upsert: false };

  const currentAgent = await Agent.findOne(searchParameter);
  if (currentAgent) {
-    const {
-      __v,
-      _id,
-      id: __id,
-      versions,
-      author: _author,
-      ...versionData
-    } = currentAgent.toObject();
+    const { __v, _id, id, versions, author, ...versionData } = currentAgent.toObject();
    const { $push, $pull, $addToSet, ...directUpdates } = updateData;

    let actionsHash = null;
@@ -382,16 +303,25 @@ const updateAgent = async (searchParameter, updateData, options = {}) => {
    }

    const shouldCreateVersion =
-      !skipVersioning &&
-      (forceVersion || Object.keys(directUpdates).length > 0 || $push || $pull || $addToSet);
+      forceVersion ||
+      (versions &&
+        versions.length > 0 &&
+        (Object.keys(directUpdates).length > 0 || $push || $pull || $addToSet));

    if (shouldCreateVersion) {
      const duplicateVersion = isDuplicateVersion(updateData, versionData, versions, actionsHash);
      if (duplicateVersion && !forceVersion) {
-        // No changes detected, return the current agent without creating a new version
-        const agentObj = currentAgent.toObject();
-        agentObj.version = versions.length;
-        return agentObj;
+        const error = new Error(
+          'Duplicate version: This would create a version identical to an existing one',
+        );
+        error.statusCode = 409;
+        error.details = {
+          duplicateVersion,
+          versionIndex: versions.findIndex(
+            (v) => JSON.stringify(duplicateVersion) === JSON.stringify(v),
+          ),
+        };
+        throw error;
      }
    }

@@ -411,7 +341,7 @@ const updateAgent = async (searchParameter, updateData, options = {}) => {
      versionEntry.updatedBy = new mongoose.Types.ObjectId(updatingUserId);
    }

-    if (shouldCreateVersion) {
+    if (shouldCreateVersion || forceVersion) {
      updateData.$push = {
        ...($push || {}),
        versions: versionEntry,
@@ -530,117 +460,12 @@ const deleteAgent = async (searchParameter) => {
  const agent = await Agent.findOneAndDelete(searchParameter);
  if (agent) {
    await removeAgentFromAllProjects(agent.id);
-    await removeAllPermissions({
-      resourceType: ResourceType.AGENT,
-      resourceId: agent._id,
-    });
  }
  return agent;
 };

-/**
- * Get agents by accessible IDs with optional cursor-based pagination.
- * @param {Object} params - The parameters for getting accessible agents.
- * @param {Array} [params.accessibleIds] - Array of agent ObjectIds the user has ACL access to.
- * @param {Object} [params.otherParams] - Additional query parameters (including author filter).
- * @param {number} [params.limit] - Number of agents to return (max 100). If not provided, returns all agents.
- * @param {string} [params.after] - Cursor for pagination - get agents after this cursor. // base64 encoded JSON string with updatedAt and _id.
- * @returns {Promise<Object>} A promise that resolves to an object containing the agents data and pagination info.
- */
-const getListAgentsByAccess = async ({
-  accessibleIds = [],
-  otherParams = {},
-  limit = null,
-  after = null,
-}) => {
-  const isPaginated = limit !== null && limit !== undefined;
-  const normalizedLimit = isPaginated ? Math.min(Math.max(1, parseInt(limit) || 20), 100) : null;
-
-  // Build base query combining ACL accessible agents with other filters
-  const baseQuery = { ...otherParams, _id: { $in: accessibleIds } };
-
-  // Add cursor condition
-  if (after) {
-    try {
-      const cursor = JSON.parse(Buffer.from(after, 'base64').toString('utf8'));
-      const { updatedAt, _id } = cursor;
-
-      const cursorCondition = {
-        $or: [
-          { updatedAt: { $lt: new Date(updatedAt) } },
-          { updatedAt: new Date(updatedAt), _id: { $gt: new mongoose.Types.ObjectId(_id) } },
-        ],
-      };
-
-      // Merge cursor condition with base query
-      if (Object.keys(baseQuery).length > 0) {
-        baseQuery.$and = [{ ...baseQuery }, cursorCondition];
-        // Remove the original conditions from baseQuery to avoid duplication
-        Object.keys(baseQuery).forEach((key) => {
-          if (key !== '$and') delete baseQuery[key];
-        });
-      } else {
-        Object.assign(baseQuery, cursorCondition);
-      }
-    } catch (error) {
-      logger.warn('Invalid cursor:', error.message);
-    }
-  }
-
-  let query = Agent.find(baseQuery, {
-    id: 1,
-    _id: 1,
-    name: 1,
-    avatar: 1,
-    author: 1,
-    projectIds: 1,
-    description: 1,
-    updatedAt: 1,
-    category: 1,
-    support_contact: 1,
-    is_promoted: 1,
-  }).sort({ updatedAt: -1, _id: 1 });
-
-  // Only apply limit if pagination is requested
-  if (isPaginated) {
-    query = query.limit(normalizedLimit + 1);
-  }
-
-  const agents = await query.lean();
-
-  const hasMore = isPaginated ? agents.length > normalizedLimit : false;
-  const data = (isPaginated ? agents.slice(0, normalizedLimit) : agents).map((agent) => {
-    if (agent.author) {
-      agent.author = agent.author.toString();
-    }
-    return agent;
-  });
-
-  // Generate next cursor only if paginated
-  let nextCursor = null;
-  if (isPaginated && hasMore && data.length > 0) {
-    const lastAgent = agents[normalizedLimit - 1];
-    nextCursor = Buffer.from(
-      JSON.stringify({
-        updatedAt: lastAgent.updatedAt.toISOString(),
-        _id: lastAgent._id.toString(),
-      }),
-    ).toString('base64');
-  }
-
-  return {
-    object: 'list',
-    data,
-    first_id: data.length > 0 ? data[0].id : null,
-    last_id: data.length > 0 ? data[data.length - 1].id : null,
-    has_more: hasMore,
-    after: nextCursor,
-  };
-};
-
 /**
 * Get all agents.
- * @deprecated Use getListAgentsByAccess for ACL-aware agent listing
 * @param {Object} searchParameter - The search parameters to find matching agents.
 * @param {string} searchParameter.author - The user ID of the agent's author.
 * @returns {Promise<Object>} A promise that resolves to an object containing the agents data and pagination info.
@@ -656,18 +481,17 @@ const getListAgents = async (searchParameter) => {
    delete globalQuery.author;
    query = { $or: [globalQuery, query] };
  }
+
  const agents = (
    await Agent.find(query, {
      id: 1,
-      _id: 1,
+      _id: 0,
      name: 1,
      avatar: 1,
      author: 1,
      projectIds: 1,
      description: 1,
-      // @deprecated - isCollaborative replaced by ACL permissions
      isCollaborative: 1,
-      category: 1,
    }).lean()
  ).map((agent) => {
    if (agent.author?.toString() !== author) {
@@ -696,7 +520,7 @@ const getListAgents = async (searchParameter) => {
 * This function also updates the corresponding projects to include or exclude the agent ID.
 *
 * @param {Object} params - Parameters for updating the agent's projects.
- * @param {IUser} params.user - Parameters for updating the agent's projects.
+ * @param {MongoUser} params.user - Parameters for updating the agent's projects.
 * @param {string} params.agentId - The ID of the agent to update.
 * @param {string[]} [params.projectIds] - Array of project IDs to add to the agent.
 * @param {string[]} [params.removeProjectIds] - Array of project IDs to remove from the agent.
@@ -729,10 +553,7 @@ const updateAgentProjects = async ({ user, agentId, projectIds, removeProjectIds
    delete updateQuery.author;
  }

-  const updatedAgent = await updateAgent(updateQuery, updateOps, {
-    updatingUserId: user.id,
-    skipVersioning: true,
-  });
+  const updatedAgent = await updateAgent(updateQuery, updateOps, { updatingUserId: user.id });
  if (updatedAgent) {
    return updatedAgent;
  }
@@ -833,14 +654,6 @@ 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
@@ -849,8 +662,8 @@ const countPromotedAgents = async () => {
 */

 module.exports = {
+  Agent,
  getAgent,
-  getAgents,
  loadAgent,
  createAgent,
  updateAgent,
@@ -859,8 +672,6 @@ module.exports = {
  revertAgentVersion,
  updateAgentProjects,
  addAgentResourceFile,
-  getListAgentsByAccess,
  removeAgentResourceFiles,
  generateActionMetadataHash,
-  countPromotedAgents,
 };
--- a/api/models/Agent.spec.js
+++ b/api/models/Agent.spec.js
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Marco Beretta	1386c38b14	refactor: correct type definitions for ToolInputSchema and error handling in searchFiles function	2025-05-31 20:53:12 +02:00
Marco Beretta	a06e999dd6	cleanup: code formatting and improve readability across multiple components	2025-05-31 20:48:23 +02:00