FastAPI MCP Server Integration with Gemini CLI

This project demonstrates how to create a FastAPI application, wrap it with an MCP (Model Context Protocol) server, and integrate it with Gemini CLI for natural language interactions.

Project Structure

.
├── app.py              # FastAPI application with REST endpoints
├── mcp_server.py       # MCP server that exposes FastAPI endpoints as tools
├── requirements.txt    # Python dependencies
├── gemini-config.json  # Gemini CLI configuration file
└── README.md          # This file

Related MCP server: FastAPI + FastMCP + Gemini Integration

Features

FastAPI Application (app.py)

• User Management: Create, read, and delete users

• Task Management: Create, read, update (complete), and delete tasks

• Calculations: Perform mathematical operations (add, subtract, multiply, divide)

• Health Check: Monitor application status

MCP Server (mcp_server.py)

Exposes 11 FastAPI endpoints as MCP tools:

• get_health - Check API health status

• get_users - Retrieve all users

• get_user - Get a specific user by ID

• create_user - Create a new user

• delete_user - Delete a user by ID

• get_tasks - Retrieve all tasks

• get_task - Get a specific task by ID

• create_task - Create a new task

• complete_task - Mark a task as completed

• delete_task - Delete a task by ID

• calculate - Perform mathematical calculations

Setup Instructions

1. Install Dependencies

pip install -r requirements.txt

2. Start the FastAPI Application

In one terminal, start the FastAPI server:

python app.py

Or using uvicorn directly:

uvicorn app:app --reload --host 0.0.0.0 --port 8000

The API will be available at http://localhost:8000

3. Install Gemini CLI

Install the Gemini CLI globally using npm:

npm install -g @google/gemini-cli@latest

4. Configure Gemini CLI

Add the MCP server to Gemini CLI configuration. The configuration file is typically located at:

• Windows: %APPDATA%\Gemini CLI\config.json

• macOS/Linux: ~/.config/gemini-cli/config.json

Or use the Gemini CLI command:

gemini mcp add fastapi-mcp-server python mcp_server.py

Alternatively, you can manually add this to your Gemini CLI config file:

{
  "mcpServers": {
    "fastapi-mcp-server": {
      "command": "python",
      "args": ["mcp_server.py"],
      "env": {
        "PYTHONPATH": "."
      }
    }
  }
}

Note: Make sure to provide the full path to mcp_server.py if running from a different directory.

5. Verify Setup

Verify that the MCP server is properly configured:

gemini mcp list

Practical Demonstration with Gemini CLI

Once everything is set up, you can interact with your FastAPI application through Gemini CLI using natural language. Here are some examples:

Example 1: Check API Health

gemini> Check the health of the FastAPI application

Gemini CLI will call the get_health tool and return the API status.

Example 2: Create a User

gemini> Create a new user named John Doe with email john@example.com and age 30

This will invoke the create_user tool with the provided parameters.

Example 3: List All Users

gemini> Show me all users

This uses the get_users tool to retrieve all users from the API.

Example 4: Create a Task

gemini> Create a task titled "Complete MCP integration" with description "Integrate MCP server with Gemini CLI"

The create_task tool will be called to create the task.

Example 5: Perform Calculation

gemini> Calculate 25 multiplied by 4

Or:

gemini> What is 100 divided by 5?

This will use the calculate tool to perform the mathematical operation.

Example 6: Complete a Task

gemini> Mark task with ID 1 as completed

This will call the complete_task tool.

Example 7: Get Task Details

gemini> Show me task number 1

Uses the get_task tool to retrieve task details.

Example 8: Delete a User

gemini> Delete user with ID 1

Invokes the delete_user tool.

Testing the API Directly

You can also test the FastAPI endpoints directly using curl or a REST client:

# Health check
curl http://localhost:8000/health

# Get all users
curl http://localhost:8000/users

# Create a user
curl -X POST http://localhost:8000/users \
  -H "Content-Type: application/json" \
  -d '{"name": "John Doe", "email": "john@example.com", "age": 30}'

# Create a task
curl -X POST http://localhost:8000/tasks \
  -H "Content-Type: application/json" \
  -d '{"title": "Test Task", "description": "This is a test task"}'

# Perform calculation
curl -X POST http://localhost:8000/calculate \
  -H "Content-Type: application/json" \
  -d '{"operation": "multiply", "a": 25, "b": 4}'

API Documentation

Once the FastAPI application is running, you can access interactive API documentation at:

• Swagger UI: http://localhost:8000/docs

• ReDoc: http://localhost:8000/redoc

Troubleshooting

MCP Server Not Found

If Gemini CLI cannot find the MCP server:

1. Ensure you're in the correct directory when starting Gemini CLI

2. Check that the path to mcp_server.py in the config is correct

3. Verify Python is in your PATH

Connection Errors

If you get connection errors:

1. Ensure the FastAPI application is running on port 8000

2. Check that the API_URL in mcp_server.py matches your FastAPI server URL

3. Verify there are no firewall restrictions

Import Errors

If you encounter import errors:

1. Make sure all dependencies are installed: pip install -r requirements.txt

2. Verify you're using the correct Python environment

3. Check that the mcp package is installed correctly

Requirements

• Python 3.8 or higher

• Node.js and npm (for Gemini CLI)

• All packages listed in requirements.txt

License

This is a sample project for educational purposes.