CrewAI Tools
StaleDescription
CrewAI Tools provides reusable integrations for the CrewAI ecosystem, including search, scraping, database access, and code execution to extend multi-agent workflows quickly.
Key Features
- Extensive tool collection including file management, web scraping, and database integrations
- MCP (Model Context Protocol) support for connecting to thousands of community MCP servers
- Two custom tool creation patterns: BaseTool subclassing and @tool decorator
- Vector database integrations with MongoDB, Qdrant, and Weaviate for semantic search
- AI-powered tools including DALL-E image generation, Vision analysis, and Stagehand browser automation
- Seamless integration with CrewAI's multi-agent framework for enhanced agent capabilities
Use Cases
💡 Extend CrewAI agents with web scraping, database query, and API integration capabilities
💡 Build custom tools for specialized domain tasks using BaseTool or decorator patterns
💡 Connect agents to MCP servers for access to community-built tools and services
💡 Create data pipelines combining vector search, file operations, and AI vision tools
Tags
Categories
Quick Start
1. Install: `pip install crewai[tools]` or `pip install crewai-tools[mcp]` for MCP support.
2. Import tools: `from crewai_tools import ScrapeWebsiteTool, SerperApiTool`.
3. Assign tools to agents: `agent = Agent(tools=[ScrapeWebsiteTool()])`.
4. For MCP: use `MCPServerAdapter` with stdio or SSE server parameters.
5. For custom tools: subclass `BaseTool` or use the `@tool` decorator.