Human MCP MCP server

The human-mcp server allows AI assistants to leverage human capabilities by sending requests to humans and receiving their responses. It acts as a bridge between AI systems and human operators, enabling AI to request humans to perform tasks that require physical senses or abilities.

Installation

Prerequisites

• Python 3.12 or higher
• uv package manager
• SQLite3

Setup Instructions

1. Clone the repository:

   git clone https://github.com/yourusername/human-mcp.git
   cd human-mcp
   

2. Create and activate a virtual environment:

   uv venv
   source .venv/bin/activate
   

3. Install dependencies:

   uv pip install .
   

4. Install the MCP server:

   task install-mcp
   

Usage

Starting the Server

To use human-mcp, you need to run both the MCP server and the Streamlit UI:

1. Connect Claude to the MCP server by configuring it with the following settings:

   "human-mcp": {
     "command": "uv",
     "args": [
       "run",
       "--with",
       "mcp[cli]",
       "mcp",
       "run",
       "$PATH_TO_REPOSITORY/human_mcp/mcp_server.py"
     ]
   }
   

2. In a second terminal, start the Streamlit UI:

   task run-streamlit
   

3. Access the Streamlit UI in your browser (typically at http://localhost:8501)

Using the Tools

The human-mcp server provides several tools that allow AI assistants to request human assistance:

• human_eye_tool: Asks a human to visually observe and describe a situation or find something
• human_hand_tool: Requests a human to perform simple physical operations
• human_mouth_tool: Instructs a human to speak specific words
• human_weather_tool: Asks a human to check and report the current weather
• human_ear_tool: Requests a human to listen and describe sounds
• human_nose_tool: Asks a human to smell and identify odors
• human_taste_tool: Requests a human to taste food and describe flavors

Workflow

1. An AI assistant (MCP client) sends a tool execution request
2. The MCP server writes the instructions to an SQLite database
3. The Streamlit app monitors the database and displays instructions to the human
4. The human enters their response via the Streamlit UI
5. The response is written back to the SQLite database
6. The MCP server reads the result and returns it to the AI assistant

When an MCP client (like Claude Desktop) sends a request, the task will appear in the Streamlit UI. The human operator can then enter their response and click the "Submit Response" button to send it back to the MCP client.

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.