The LLM was not seeing MCP tools when using standalone agent execution
(without Crew) because the prepare_tools function in agent/utils.py
did not inject MCP tools.
Changes:
- Modified prepare_tools() in agent/utils.py to inject MCP tools when
agent.mcps is configured, with graceful error handling
- Fixed Agent.kickoff_async() to inject MCP tools like kickoff() does
- Added comprehensive tests for MCP tool injection in prepare_tools
The fix ensures MCP tools are visible to the LLM in both:
1. Standalone agent execution via execute_task/aexecute_task
2. Async agent execution via kickoff_async
Co-Authored-By: João <joao@crewai.com>
- Enhanced the MCP tool execution in both synchronous and asynchronous contexts by utilizing for better event loop management.
- Updated error handling to provide clearer messages for connection issues and task cancellations.
- Added tests to validate MCP tool execution in both sync and async scenarios, ensuring robust functionality across different contexts.
* WIP transport support mcp
* refactor: streamline MCP tool loading and error handling
* linted
* Self type from typing with typing_extensions in MCP transport modules
* added tests for mcp setup
* added tests for mcp setup
* docs: enhance MCP overview with detailed integration examples and structured configurations
* feat: implement MCP event handling and logging in event listener and client
- Added MCP event types and handlers for connection and tool execution events.
- Enhanced MCPClient to emit events on connection status and tool execution.
- Updated ConsoleFormatter to handle MCP event logging.
- Introduced new MCP event types for better integration and monitoring.
