CrewAI (Near Intents) MCP server

The Crew AI MCP Server provides AI agent and task management capabilities through the CrewAI framework, allowing you to create and manage AI agents that can perform various tasks as part of a coordinated crew.

Installation

To install the Crew AI MCP Server, follow these steps:

1. Clone or fork the repository:

   git clone https://github.com/your-username/crew-ai-mcp-server.git
   cd crew-ai-mcp-server
   

2. Run the setup script:

   ./crew.sh
   

This setup script will:

• Install required Python dependencies
• Configure the MCP settings file for your system
• Set up the correct paths automatically

Configuration

Before using the server, you need to set your OpenAI API key:

export OPENAI_API_KEY="your-api-key"

System Requirements

To use the Crew AI MCP Server, you'll need:

• Python 3.8 or higher
• jq command-line tool (for setup script)
• VSCode with Roo Cline extension installed

The server supports the following platforms:

• macOS
• Linux
• Windows (via Git Bash)

Using the Server

The server provides three main tools that can be used to create AI workflows:

Create an Agent

To create an AI agent, use the following JSON format:

{
    "method": "call_tool",
    "params": {
        "name": "create_agent",
        "arguments": {
            "role": "researcher",
            "goal": "Research and analyze information effectively",
            "backstory": "An experienced research analyst"
        }
    }
}

Create a Task

To create a task for an agent to perform, use:

{
    "method": "call_tool",
    "params": {
        "name": "create_task",
        "arguments": {
            "description": "Analyze recent market trends",
            "agent": "researcher",
            "expected_output": "A detailed analysis report"
        }
    }
}

Create and Run a Crew

To create and run a crew (a collection of agents working on tasks), use:

{
    "method": "call_tool",
    "params": {
        "name": "create_crew",
        "arguments": {
            "agents": ["researcher"],
            "tasks": ["Analyze recent market trends"],
            "verbose": true
        }
    }
}

Complete Workflow Example

You can create and run a complete workflow with the following command:

(echo '{"method": "call_tool", "params": {"name": "create_agent", "arguments": {"role": "researcher", "goal": "Research and analyze information effectively", "backstory": "An experienced research analyst"}}}'; echo '{"method": "call_tool", "params": {"name": "create_task", "arguments": {"description": "Analyze recent market trends", "agent": "researcher", "expected_output": "A detailed analysis report"}}}'; echo '{"method": "call_tool", "params": {"name": "create_crew", "arguments": {"agents": ["researcher"], "tasks": ["Analyze recent market trends"], "verbose": true}}}') | python3 src/crew_server.py

This example:

1. Creates a researcher agent
2. Creates an analysis task
3. Assembles a crew with the agent and task
4. Runs the workflow

Troubleshooting

If you encounter any issues:

• Ensure your OpenAI API key is set correctly
• Check that all dependencies are installed (pip install -r requirements.txt)
• Verify the MCP settings file exists and has the correct configuration
• Make sure the server path in the MCP settings matches your actual file location

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.