python-sdk
Python SDK for building MCP servers and clients. It helps apps expose tools and resources to LLM systems through a standard protocol.
Share on XLicense: MIT
Overview
This project is the official Python SDK for Model Context Protocol (MCP). It gives developers a standard way to build MCP servers that expose tools, resources, and prompts, and MCP clients that connect to those services. The SDK supports stdio, Streamable HTTP, and SSE transports, and is designed for Python 3.10+ projects that need to integrate LLM applications with external data and functionality in a consistent way.
Key features
- Build MCP servers that expose tools, resources, and prompts
- Build MCP clients that connect to MCP servers
- Use stdio, Streamable HTTP, and SSE transports
- Work with Python 3.10+ and the mcp CLI for local development
Best for
Developers building Python integrations for LLM applications and MCP-compatible services. It is a fit when you need a standard server/client setup without writing protocol handling yourself.
- Upstream
- modelcontextprotocol/python-sdk
- Fork on GitHub
- Guo-astro/python-sdk
- Upstream stars
- 24k
- Category
- AI agents and LLM tools
- License
- MIT
- Forked
- 2026-10-03
- Sync status
- In syncLast synced 2026-10-05
More in AI agents and LLM tools
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA, and ROCm. It helps run capable open-weight models on consumer hardware without relying on a cloud service.
Forked 2026-10-03Last synced 2026-10-05License: MITAI agents and LLM toolsGitHub
Strata lets you run a large local AI model on a gaming PC so work stays on your hardware. It helps with chats, code work, image input, and app integration without needing a server.
Forked 2026-10-03Last synced 2026-10-05License: MITAI agents and LLM toolsGitHub
Strands SDK helps teams build and run production AI agents in Python or TypeScript with any model and cloud, without writing a custom agent loop.
Forked 2026-10-02Last synced 2026-10-05License: Apache-2.0AI agents and LLM toolsGitHub
Trailblaze lets coding agents drive real devices and web apps with natural-language steps, then saves sessions as replayable trails for CI tests. It keeps flows repeatable without LLMs at replay time.
Forked 2026-10-02Last synced 2026-10-05License: Apache-2.0AI agents and LLM toolsGitHub