home / mcp / manusmcp mcp server

ManusMCP MCP Server

Orchestrates specialized AI agents via MCP to collaborate on complex tasks.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "mantrakp04-manusmcp": {
      "command": "bunx",
      "args": [
        "@modelcontextprotocol/inspector",
        "uv",
        "run",
        "manusmcp"
      ]
    }
  }
}

ManusMCP is a framework for orchestrating specialized AI agents that work together to accomplish complex tasks by providing a shared context and communication layer called the Model Context Protocol (MCP). This setup enables you to deploy AI team members with focused capabilities, coordinating them to tackle projects more efficiently and consistently.

How to use

You can run ManusMCP locally or in a container environment and then connect an MCP client to manage and observe the agents. The setup focuses on practical usage patterns: launching the server, loading agent workflows, and guiding the agents through tasks using the built-in MCP flow integration.

How to install

Prerequisites you need before installing ManusMCP:

Git is installed on your system
Node.js and npm or a compatible runtime are available
bun is available if you plan to use the bun runtime
Docker is installed if you want to run via docker-compose

Concrete installation steps you can follow:

# Clone the project
git clone https://github.com/mantrakp04/manusmcp.git
cd manusmcp/pytools

# Install dependencies
npm i -g bun
bash flowise.sh

Additional sections

MCP server usage involves launching the server and optionally inspecting or managing it via an inspector tool. A common workflow is to run the inspector in a local, self-contained environment to observe MCP behavior and validate agent interactions.

# Optional: start the MCP server inspector (if you want to explore MCP interactions locally)
bunx @modelcontextprotocol/inspector uv run manusmcp