mcp-use
by syw2014
Overview
A full-stack framework for building Model Context Protocol (MCP) servers, clients, and AI agents in Python and TypeScript.
Installation
npm run devEnvironment Variables
- GITHUB_TOKEN
- MCP_USE_API_KEY
- LANGFUSE_PUBLIC_KEY
- LANGFUSE_SECRET_KEY
- MCP_USE_AGENT_ENV
- PORT
- HOST
- MCP_URL
Security Notes
The framework is designed to execute arbitrary commands (via `npx` or direct process spawning) on the host machine as part of its `MCPClient` configuration (e.g., `command` and `args` in server configs). This functionality, while core to dynamic server loading, presents a significant command injection and sandbox escape risk if the client configuration or inputs to an AI agent are not strictly validated, sanitized, and run within a highly isolated environment. The `inspector` also includes proxy capabilities that could be misused if not properly secured. Hardcoded secrets are not evident; environment variables are used for sensitive information.
Similar Servers
fastmcp
FastMCP is an ergonomic interface for the Model Context Protocol (MCP), providing a comprehensive framework for building and interacting with AI agents, tools, resources, and prompts across various transports and authentication methods.
mcp-use
A comprehensive framework for building full-stack Model Context Protocol (MCP) applications, including AI agents, MCP servers with UI widgets, and integrated debugging tools in both Python and TypeScript.
AgentUp
A developer-first framework for building, deploying, and managing secure, scalable, and configurable AI agents, supporting various agent types (reactive, iterative) and the Model-Context Protocol (MCP) for seamless interactions.
boilerplate-mcp-server
Provides a production-ready foundation for developing custom Model Context Protocol (MCP) servers in TypeScript to connect AI assistants with external APIs and data sources, exemplified by an IP geolocation tool.