Mcacp
MCACP is an MCP server that can bridge any MCP client to any ACP coding agent, enabling the control of a multi-agent system from a single chat window, supporting agent installation, session management, and permission control.
rating : 2 points
downloads : 6.9K
What is MCACP?
MCACP is a Model Context Protocol (MCP) server that establishes a connection between your MCP host and Agent Client Protocol (ACP) coding agents. Simply put, it enables you to run, manage, and interact with various coding agents in any MCP-supported chat interface without installing dedicated IDE plugins for each agent.How to use MCACP?
Using MCACP is very simple: First, install the MCACP server, and then add it to your MCP client configuration. After that, you can directly search for, install, and run various ACP agents from the chat interface, such as Claude Code, Gemini CLI, Codex, etc.Use cases
MCACP is particularly suitable for the following scenarios: 1) Want to use multiple coding agents in the same interface; 2) Need to compare the performance of different agents in parallel; 3) Build a workflow for multi-agent collaboration; 4) Unifiedly manage agent permissions and sessions in a team.Main features
Multi-agent support
Supports running and managing multiple ACP coding agents, including Claude Code, Gemini CLI, Codex, Auggie, Copilot, etc., all operated in the same interface.
Cross-platform compatibility
Compatible with all MCP-supported clients, including Claude Code, Zed, VS Code, etc., without installing different plugins for different platforms.
Flexible permission control
Provides multiple permission policies: allow all, deny all, interactive inquiry, or operator mode. Different permission levels can be set for each session as needed.
Session persistence
Supports saving and restoring session states. Even if the client is closed or the server is restarted, you can continue working from where you left off.
Agent discovery and installation
Built-in agent registry search function, allowing you to browse, install, and update various ACP agents directly from the chat interface.
Parallel comparison
Can run multiple agents simultaneously to handle the same task, facilitating the comparison of the performance and output results of different agents.
Advantages
Unified management interface: Manage all coding agents in a single chat window without switching between different tools
Flexible agent combination: Freely combine different agents to build a solution that best suits your workflow
Fine-grained permission control: Set independent permission policies for each session to ensure security and controllability
Easy to expand: Supports custom agent configuration and can integrate self-developed agent tools
Cost-effective: No need to purchase separate IDE plugin licenses for each agent
Limitations
Depends on MCP support: Requires the client to support the MCP protocol to use
Configuration complexity: Initial setup requires certain technical knowledge to configure the MCP client
Performance overhead: Running multiple agents may consume more system resources
Learning curve: Need to understand the basic concepts of MCP and ACP to fully utilize all functions
How to use
Install MCACP
Install the MCACP server globally via npm, or run it directly using npx.
Configure the MCP client
Add MCACP to your MCP client configuration. Here are configuration examples for Claude Code and Zed.
Search for and install agents
Use the registry_search command in the chat interface to search for available agents, and then use agent_install to install them.
Initialize and create a session
Initialize the agent process, create a new working session, and set the permission policy.
Start using the agent
Send a prompt to the agent to start the coding task.
Usage examples
Parallel code review
Use two different agents to review the same piece of code simultaneously and compare their feedback and suggestions.
Multi-stage development pipeline
Build a three-stage development pipeline: One agent writes the code, the second agent reviews the code, and the third agent writes the tests.
Technology stack migration
Use agents with different expertise to assist in migrating a project from one technology stack to another.
Frequently Asked Questions
Which MCP clients does MCACP support?
How to manage agent permissions?
Where is the session data stored?
How many agents can be run simultaneously?
How to add a custom agent?
Is MCACP free?
Related resources
MCACP GitHub repository
Source code, issue tracking, and contribution guidelines for MCACP
Model Context Protocol official documentation
Complete specification, tutorials, and implementation guidelines for the MCP protocol
Agent Client Protocol documentation
Detailed specification and agent development guidelines for the ACP protocol
Claude Code MCP guide
Configuration and usage tutorials for using the MCP server in Claude Code
Zed editor MCP support
MCP integration documentation and configuration examples for the Zed editor

Gitlab MCP Server
Certified
The GitLab MCP server is a project based on the Model Context Protocol that provides a comprehensive toolset for interacting with GitLab accounts, including code review, merge request management, CI/CD configuration, and other functions.
TypeScript
24.7K
4.3 points

Markdownify MCP
Markdownify is a multi-functional file conversion service that supports converting multiple formats such as PDFs, images, audio, and web page content into Markdown format.
TypeScript
34.6K
5 points

Duckduckgo MCP Server
Certified
The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.
Python
72.6K
4.3 points

Notion Api MCP
Certified
A Python-based MCP Server that provides advanced to-do list management and content organization functions through the Notion API, enabling seamless integration between AI models and Notion.
Python
20.5K
4.5 points

Figma Context MCP
Framelink Figma MCP Server is a server that provides access to Figma design data for AI programming tools (such as Cursor). By simplifying the Figma API response, it helps AI more accurately achieve one - click conversion from design to code.
TypeScript
63.8K
4.5 points

Unity
Certified
UnityMCP is a Unity editor plugin that implements the Model Context Protocol (MCP), providing seamless integration between Unity and AI assistants, including real - time state monitoring, remote command execution, and log functions.
C#
31.5K
5 points

Gmail MCP Server
A Gmail automatic authentication MCP server designed for Claude Desktop, supporting Gmail management through natural language interaction, including complete functions such as sending emails, label management, and batch operations.
TypeScript
21.1K
4.5 points

Minimax MCP Server
The MiniMax Model Context Protocol (MCP) is an official server that supports interaction with powerful text-to-speech, video/image generation APIs, and is suitable for various client tools such as Claude Desktop and Cursor.
Python
49.5K
4.8 points




