If you are looking for the TypeScript docs, they are here.
What is MCP?
The Model Context Protocol (MCP) provides a standardized way to connect LLMs to tools and data sources. Think of it as an API specifically designed for LLM interactions. MCP servers can:- Expose data through Resources (similar to GET endpoints for loading information into LLM context)
- Provide functionality through Tools (similar to POST endpoints for executing operations)
- Define interaction patterns through Prompts (reusable templates for LLM interactions)
Why mcp-use?
Complete Vertical Stack
Build everything from AI agents to MCP servers. Create the full MCP ecosystem without needing multiple libraries.Production Ready
Includes streaming, multi-server support, authentication, observability with Langfuse, and built-in security features.Python Native
Clean Pythonic API with full type hints, async/await support, and seamless integration with popular frameworks.Developer Experience
Built-in MCP Inspector for debugging, comprehensive documentation, and easy integration with LangChain, OpenAI, Anthropic, and Google.Next Steps
Quickstart
Get started with MCP servers, clients, and agents in minutes
Installation
Install mcp-use and set up your development environment
MCP Server
Build custom MCP servers with tools, resources, and prompts
MCP Client
Connect to any MCP server programmatically
MCP Agent
Create AI agents with tool calling capabilities
Inspector
Debug and test MCP servers interactively