--- id: mcpadapt version: "0.1.20" license: MIT license_treatment: permissive maintenance: aging --- # mcpadapt — Adapt MCP servers to many agentic framework. License: permissive · Maintenance: aging · Downloads: 217.8K/mo ## What it is and what it does MCPAdapt is a bridge layer that converts MCP servers—a protocol introduced by Anthropic for exposing tools and resources—into native tool objects for popular agentic frameworks. It lets you take any of the 650+ MCP servers available in the ecosystem and use them directly within Smolagents, Langchain, CrewAI, or Google GenAI without writing framework-specific glue code. The package handles both local MCP servers (via stdio) and remote ones (via SSE), and supports both synchronous and asynchronous execution where the underlying framework permits. It depends on jsonref, mcp, pydantic, and python-dotenv, and is designed to be extended with new framework adapters through a simple plugin interface. Use it for: - Integrate a local MCP server into a Smolagents workflow to give an AI agent access to custom tools without rewriting them. - Connect multiple MCP servers to a Langchain agent to provide diverse data sources and capabilities in a single tool collection. - Adapt a remote MCP server (via SSE) to CrewAI for use in multi-agent workflows. - Expose third-party MCP tools from smithery.ai or glama.ai to your agentic framework without manual tool definition. - Build a custom adapter for an unsupported agentic framework by implementing the ToolAdapter interface. ## Worth the install? AI-flagged interpretation of the facts above — verify before relying. Adapts MCP (Model Context Protocol) servers into tools for agentic frameworks like Smolagents, Langchain, and CrewAI, exposing their capabilities to AI agents. Yes, if you are building agents with Smolagents, Langchain, CrewAI, or Google GenAI and want to leverage the MCP ecosystem. The package is low-friction to install and has no known vulnerabilities. However, note that maintenance is aging (last update October 2025, no commits since); evaluate whether the supported frameworks and adapters match your stack before committing. The security warning about SSE servers in the documentation should be taken seriously. ## Install pip install mcpadapt uv add mcpadapt poetry add mcpadapt ## Installing mcpadapt Before you install: Low install friction with a pure-Python wheel and four runtime dependencies. Maintenance status is aging—the project was first released in January 2025 and last updated in October 2025, with no commits since then, though the repository remains active with 423 stars. License in practice: MIT license permits commercial and private use with minimal restrictions; you must retain the license notice in distributions. Quickstart: from mcp import StdioServerParameters from mcpadapt.core import MCPAdapt from mcpadapt.smolagents_adapter import SmolAgentsAdapter with MCPAdapt( StdioServerParameters(command="uv", args=["run", "src/echo.py"]), SmolAgentsAdapter(), ) as tools: # use tools in your agentic workflow Requires Python 3.12 or later; MCP server command must be executable and available in your environment. Verify before relying: - Stability and maturity of the adapter implementations for each supported framework. - Performance characteristics when connecting to multiple MCP servers simultaneously. - Security audit status of SSE (Server-Sent Events) remote server connections. ## Package facts - License: MIT (permissive) - Python support: supports_current - Install friction: low - Maintenance: aging - Downloads: 217.8K/month (top 15,000 on PyPI) - Known vulnerabilities: none known ## Tags MCP server integration agentic framework, model context protocol adapter, agent tool integration MCP, connect MCP servers to agents, agentic framework tool collection, MCP tools for langchain smolagents, protocol adapter agent framework, agent-framework-integration, mcp-protocol, tool-adapter [View on SkillFed](https://skillfed.io/packages/mcpadapt) · [View on PyPI](https://pypi.org/project/mcpadapt/)