- Introduced DedupHITLToolCallsMiddleware to prevent duplicate HITL tool calls within a single LLM response, ensuring only the first occurrence of each tool call is retained.
- Updated the create_surfsense_deep_agent function to include the new middleware, enhancing the efficiency of tool interactions.
- Added a new middleware file for better organization and maintainability of the codebase.
- Added a guideline to ensure that each tool (Gmail, Google Calendar, Google Drive, Linear, Notion) is called only once per user request.
- Updated documentation to clarify that the system will automatically select the most relevant match when multiple items share the same title or subject, enhancing user experience and preventing redundant calls.
- Improved the update_gmail_draft tool to allow users to review and edit draft content before applying changes.
- Added logic to generate draft body content based on user requests and conversation context.
- Implemented fetching of existing draft body to facilitate user edits in the approval card.
- Updated UI components to support displaying and editing existing draft content, enhancing user experience.
- Introduced a new tool to update existing Gmail drafts, allowing users to modify draft content, recipients, and subject lines.
- Updated the Gmail tools registry to include the new update_gmail_draft tool.
- Enhanced the GmailKBSyncService to support draft ID handling during synchronization.
- Added UI components for the update draft functionality in the web application, improving user interaction with Gmail drafts.
- Implemented logic to persist authentication expiration status for connectors when insufficient permissions are detected, improving error handling and user experience.
- Updated messages to guide users to re-authenticate in connector settings for Gmail, Google Calendar, and Google Drive tools.
- Added InsufficientPermissionsResult type and corresponding UI components to display permission-related messages consistently across Gmail and Google Calendar tools.
- Added `sync_after_create` methods in `LinearKBSyncService` and `NotionKBSyncService` to handle synchronization of newly created issues and pages with the knowledge base.
- Enhanced the `create_issue.py` and `create_page.py` tools to provide user feedback on the success of the knowledge base update, indicating whether the content has been synced or will be added in the next scheduled sync.
- Improved error handling during synchronization to log failures and manage duplicate document scenarios effectively.
- Implemented checks for expired authentication in the Notion page creation and deletion tools, returning appropriate error messages for re-authentication.
- Updated the Notion tool metadata service to track account health and persist authentication status, improving error handling and user feedback during operations.
- Added a new GoogleDriveKBSyncService to handle synchronization of newly created Google Drive files with the knowledge base.
- Enhanced the create_file.py tool to include feedback on the success of the knowledge base update, informing users if their file has been added or will be synced later.
- Updated the Google Drive tool metadata service to include parent folder information for improved file organization.
- Modified the UI components to support selection of parent folders during file creation, enhancing user experience and file management.
- Added checks for expired authentication in Google Drive file creation and deletion tools, returning appropriate error messages for re-authentication.
- Updated the Google Drive tool metadata service to track account health and persist authentication status.
- Enhanced UI components to display authentication errors and differentiate between valid and expired accounts, improving user experience during file operations.
- Introduced a mapping of native Google document types to their legacy Composio equivalents, ensuring seamless search and indexing for both types.
- Updated relevant components to utilize the new mapping, enhancing the consistency of document type handling across the application.
- Improved search functionality to transparently include legacy types, maintaining accessibility for older documents until re-indexed.
- Implemented fallback logic in Google Calendar, Drive, and Gmail indexers to handle legacy Composio document types, ensuring smooth migration to native types.
- Updated document indexing functions to check for existing documents using both primary and legacy hashes, improving data integrity during indexing.
- Added re-authentication endpoints for Linear and Notion connectors to handle expired authentication.
- Enhanced error handling in issue creation, deletion, and update tools to return appropriate authentication error messages.
- Updated UI components to display authentication status and guide users on re-authentication steps.
- Improved account health checks to ensure valid tokens are used for operations.
- modify the `web_search` tool that integrates multiple live search connectors (Tavily, Linkup, Baidu) alongside the SearXNG platform for real-time information retrieval.
- Updated comments to reflect the separation of local knowledge base searches from web searches.
- Improved handling of connector types and search logic to ensure accurate routing of queries to the appropriate tools.
- Refactored existing code to streamline the search process and enhance performance.
- Deleted the migration script for adding web search columns to the database.
- Removed web search related fields from the SearchSpace model and schemas.
- Eliminated web search health check endpoint and associated service logic.
- Updated frontend components to remove web search settings management.
- Cleaned up references to SearXNG API in various modules, transitioning to a platform service approach.
- Added endpoint to list agent tools with metadata, excluding hidden tools.
- Updated NewChatRequest and RegenerateRequest schemas to include disabled tools.
- Integrated disabled tools management in the NewChatPage and Composer components.
- Improved tool instructions and visibility in the system prompt.
- Refactored tool registration to support hidden tools and default enabled states.
- Enhanced document chunk creation to handle strict zip behavior.
- Cleaned up imports and formatting across various files for consistency.
- Introduced a shielded async session context manager to ensure safe session closure during cancellations.
- Updated various database operations to utilize the new shielded session, preventing orphaned connections.
- Added environment variables to optimize glibc memory management, improving overall application performance.
- Implemented a function to trim the native heap, allowing for better memory reclamation on Linux systems.
- Introduced a mechanism to identify degenerate queries that lack meaningful search signals, improving search accuracy.
- Implemented a fallback method for browsing recent documents when queries are degenerate, ensuring relevant results are returned.
- Added limits on the number of chunks fetched per document to optimize performance and prevent excessive data loading.
- Updated the ConnectorService to allow for reusable query embeddings, enhancing efficiency in search operations.
- Enhanced LLM router service to support context window fallbacks, improving robustness during context window limitations.
- Improved in-memory rate limiting by evicting timestamps outside the current window and cleaning up empty keys.
- Updated LLM router service to cache context profiles and avoid redundant computations.
- Introduced cache eviction logic for MCP tools and sandbox instances to manage memory usage effectively.
- Added garbage collection triggers in chat streaming functions to reclaim resources promptly.
- improved search_knowledgebase_tool
- Added new endpoint to batch-fetch comments for multiple messages, reducing the number of API calls.
- Introduced CommentBatchRequest and CommentBatchResponse schemas for handling batch requests and responses.
- Updated chat_comments_service to validate message existence and permissions before fetching comments.
- Enhanced frontend with useBatchCommentsPreload hook to optimize comment loading for assistant messages.
- Introduced RequestPerfMiddleware to log request performance metrics, including slow request thresholds.
- Updated various services and retrievers to utilize the new performance logging utility for better tracking of execution times.
- Enhanced existing methods with detailed performance logs for operations such as embedding, searching, and indexing.
- Removed deprecated logging setup in stream_new_chat and replaced it with the new performance logger.
- Introduced dynamic character budget calculation for document formatting based on model's context window.
- Updated `format_documents_for_context` to respect character limits and improve output quality.
- Added `max_input_tokens` parameter to various functions to facilitate context-aware processing.
- Enhanced error handling for context overflow in LLM router service.
- Replaced direct embedding calls with a utility function across various components to streamline embedding logic.
- Added enable_summary flag to several models and routes to control summary generation behavior.
- Introduced slow callback logging in FastAPI to identify blocking calls.
- Added performance logging for agent creation and tool loading processes.
- Implemented caching for MCP tools to reduce redundant server calls.
- Enhanced sandbox management with in-process caching for improved efficiency.
- Refactored several functions for better readability and performance tracking.
- Updated tests to ensure proper functionality of new features and optimizations.
