MCP Agent Tool Adapter

v1.0.0Developer Toolsstable

Transforms MCP tools into collaborative, reasoning agents using modern agent frameworks.

agentsgoogle-adklangchainlanggraphmcp
Share:
8
Stars
0
Downloads
0
Weekly
0/5

What is MCP Agent Tool Adapter?

MCP Agent Tool Adapter is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to transforms mcp tools into collaborative, reasoning agents using modern agent frameworks.

Transforms MCP tools into collaborative, reasoning agents using modern agent frameworks.

This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Transforms MCP tools into collaborative, reasoning agents us

Use Cases

Convert MCP tools into collaborative reasoning agents.
Integrate with LangChain and LangGraph frameworks.
serkanyasr

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedAug 21, 2025
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx mcp-agent-tool-adapter

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use MCP Agent Tool Adapter

The MCP Agent Tool Adapter is a Python framework that wraps standard MCP tool servers and re-exposes them as autonomous reasoning agents. It supports two agent backends — Google ADK (with a FastAPI or CLI interface) and LangGraph (using the ReAct pattern with streaming) — so you can compose multiple MCP tool servers into a single collaborative agent pipeline.

Prerequisites

  • Python 3.10 or later with pip installed
  • Git to clone the repository
  • At least one running MCP tool server to connect (e.g., filesystem server via npx)
  • Optional: Google ADK credentials if using the ADK backend
  • Optional: OpenAI or compatible LLM API key for the LangGraph ReAct agent
1

Clone the repository

Clone the mcp-agent-tool-adapter repository to your local machine.

git clone https://github.com/serkanyasr/mcp-agent-tool-adapter.git
cd mcp-agent-tool-adapter
2

Install Python dependencies

Install all required packages from the requirements file.

pip install -r requirements.txt
3

Configure mcp_config.json

Create or edit mcp_config.json in the project root to list the MCP tool servers you want to include. Each entry specifies the command and arguments to launch that server.

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["@modelcontextprotocol/server-filesystem", "/tmp"]
    }
  }
}
4

Run with the Google ADK backend

Launch the adapter using the Google ADK client. This starts a FastAPI server or CLI interface that exposes the combined agent.

python app_client_adk.py
5

Run with the LangGraph ReAct backend

Alternatively, launch the LangGraph agent for streaming ReAct-style reasoning across your configured MCP tools.

python app_client_langgraph.py

MCP Agent Tool Adapter Examples

Client configuration

Example mcp_config.json wiring up a filesystem MCP server as a tool source for the adapter.

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["@modelcontextprotocol/server-filesystem", "/home/user/data"]
    },
    "brave-search": {
      "command": "npx",
      "args": ["@modelcontextprotocol/server-brave-search"],
      "env": {
        "BRAVE_API_KEY": "your-api-key"
      }
    }
  }
}

Prompts to try

Interact with the running agent over its ADK or LangGraph interface.

- "Search the web for the latest LangGraph release notes and save a summary to /tmp/notes.txt."
- "List files in the data directory and analyze the contents of the largest CSV."
- "Use all available tools to research a topic and write a structured report."
- "Coordinate a multi-step task: first search for information, then store results locally."

Troubleshooting MCP Agent Tool Adapter

ModuleNotFoundError when starting app_client_adk.py or app_client_langgraph.py

Ensure you ran `pip install -r requirements.txt` inside the correct virtual environment. Activate your venv first: `source venv/bin/activate` then reinstall.

MCP tool server fails to launch from mcp_config.json

Verify the command and args in mcp_config.json are correct and the referenced binary (e.g., npx) is on your PATH. Test the server launch command manually in a terminal first.

LangGraph agent hangs or produces no output

Check that your LLM API key environment variable is set correctly. The ReAct agent requires an LLM endpoint. Export OPENAI_API_KEY or the relevant key before starting.

Frequently Asked Questions about MCP Agent Tool Adapter

What is MCP Agent Tool Adapter?

MCP Agent Tool Adapter is a Model Context Protocol (MCP) server that transforms mcp tools into collaborative, reasoning agents using modern agent frameworks. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install MCP Agent Tool Adapter?

Follow the installation instructions on the MCP Agent Tool Adapter GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with MCP Agent Tool Adapter?

MCP Agent Tool Adapter works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is MCP Agent Tool Adapter free to use?

Yes, MCP Agent Tool Adapter is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Developer Tools MCP Servers

Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "mcp-agent-tool-adapter": { "command": "npx", "args": ["-y", "mcp-agent-tool-adapter"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use MCP Agent Tool Adapter?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides