ArcadeAI/arcade-mcp
Python framework for building Model Context Protocol servers and tools that extend agent capabilities.

Arcade MCP is an open-source framework for building Model Context Protocol servers and tools, designed to add custom capabilities to AI agents. It provides a decorator-based API covering the full MCP specification including tools, resources, prompts, and sampling. The framework supports authorized tool calling with OAuth handling and integrates with popular agent frameworks including LangChain, Mastra, Pydantic AI, CrewAI, and OpenAI Agents. It powers over 7,500 prebuilt tools across Arcade.dev and includes CLI tooling for evaluation and deployment.
Frequently asked
- What is ArcadeAI/arcade-mcp?
- Python framework for building Model Context Protocol servers and tools that extend agent capabilities.
- Is arcade-mcp open source?
- Yes — ArcadeAI/arcade-mcp is open source, released under the MIT license.
- What language is arcade-mcp written in?
- ArcadeAI/arcade-mcp is primarily written in Python.
- How popular is arcade-mcp?
- ArcadeAI/arcade-mcp has 976 stars on GitHub.
- Where can I find arcade-mcp?
- ArcadeAI/arcade-mcp is on GitHub at https://github.com/ArcadeAI/arcade-mcp.