ACP to MCP Adapter logo

ACP to MCP Adapter

by i-am-bee

The ACP to MCP Adapter bridges Agent Communication Protocol (ACP) and Model Context Protocol (MCP) ecosystems. It allows MCP applications to discover and interact with ACP agents as resources.

View on GitHub

Last updated: N/A

What is ACP to MCP Adapter?

The ACP to MCP Adapter is a lightweight server that connects ACP agents to MCP applications, enabling seamless interaction between the two AI ecosystems.

How to use ACP to MCP Adapter?

To use the adapter, run it with the ACP server URL, then configure your MCP client application (e.g., Claude Desktop) to connect to the adapter. Invoke ACP agents through the MCP client using the 'run_agent' tool or by attaching them as MCP resources.

Key features of ACP to MCP Adapter

  • Makes ACP agents discoverable as MCP resources

  • Exposes ACP agent runs as MCP tools

  • Bridges two agent ecosystems with minimal configuration

  • Supports Stdio transport

Use cases of ACP to MCP Adapter

  • Connecting ACP agents to Claude Desktop

  • Enabling MCP applications to utilize ACP agents

  • Integrating different AI ecosystems

  • Running ACP agents from MCP environments

FAQ from ACP to MCP Adapter

What is ACP?

Agent Communication Protocol, used for agent-to-agent communication.

What is MCP?

Model Context Protocol, used for connecting AI models to external tools.

What are the requirements to run the adapter?

Python 3.11+, acp-sdk, mcp, a running ACP server, and an MCP client application.

How do I connect Claude Desktop to the adapter?

Edit the Claude Desktop configuration file (claude_desktop_config.json) to include the acp-mcp server details.

How do I invoke an ACP agent from Claude Desktop?

Use the 'run_agent' tool or attach the agent as an MCP resource and then interact with it.