Fast

You run a fast, API-driven MCP server that integrates with Google Gemini to manage tasks, analyze data, and automate workflows. This server exposes a RESTful interface for task CRUD operations, analytics exports, and AI-powered responses, while providing a CLI for quick local interactions and a Gemini integration for natural language and code assistance.

How to use

Start by launching the server through your MCP client workflow. You can interact with the HTTP API to manage tasks, fetch statistics, and export data, or use the built‑in CLI for quick operations. Gemini integration enables AI-powered prompts and code reviews, with context maintained across interactions.

Common usage patterns include creating tasks, updating their status, listing tasks with optional filters, and exporting data to CSV. You can also query AI guidance or code examples via the Gemini integration. All interactions are designed to be straightforward and scriptable for automation.

Representative workflows you can perform through the MCP client or direct API calls include: creating a task with a title, description, and priority; updating a task’s status; listing tasks with a status filter; obtaining task statistics; and exporting tasks to CSV. For AI-assisted needs, send a Gemini prompt to obtain best practices, code reviews, or documentation help. Remember that the server runs on port 5000 and exposes a Swagger UI for API exploration.

How to install

Prerequisites you need before installing: Python 3.8 or higher, Google Gemini CLI installed and configured, curl for CLI operations, and Git for cloning.

git clone <repository-url>
cd FAST_MCP

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

pip install -r requirements.txt

# Install Google Gemini CLI as prompted by its official setup
# Configure the Gemini API key after installation, e.g. gemini config set api_key YOUR_API_KEY

Next, start the MCP server through the provided CLI utility. The server will run on http://localhost:5000 and expose the REST API and interactive docs.

Configuration and deployment notes

Server port and host: the server runs by default on port 5000 and binds to all interfaces (0.0.0.0). You can access the API docs via Swagger UI at /docs and the ReDoc interface at /redoc.

Data and logs: the server uses a local JSON file for task data and produces JSON-formatted operation logs for auditing and debugging. Ensure your environment has appropriate file permissions for data and log files.

Gemini integration: prompts sent to the Gemini CLI are logged and handled with a 30-second timeout to prevent hanging requests. Ensure the Gemini API key is configured correctly to enable AI responses.

Developer notes

APIs are documented through the Swagger UI once the server is running, enabling you to discover endpoints and try them directly in your browser.

The CLI provides commands to manage tasks and interact with Gemini. Use the CLI to list tasks, create tasks, update statuses, search tasks, export CSVs, and query Gemini for prompts like code reviews or documentation help.

Security and best practices

Protect access to the server with appropriate authentication in production. Consider adding API keys, OAuth, or other access controls, plus rate limiting and monitoring to guard against misuse.

Troubleshooting

If the server won’t start, check that port 5000 is free, the Python virtual environment is activated, and server logs are available for debugging.

If Gemini integration fails, verify the Gemini CLI is installed, the API key is configured, and there are no network timeouts preventing CLI communication.

For CLI-related issues, ensure mcp_cli.sh has execute permissions and follow the CLI’s help prompts to diagnose command usage.

API and docs access

Once the server is running, you can navigate to the interactive API docs at http://localhost:5000/docs for Swagger UI and http://localhost:5000/redoc for ReDoc documentation.

Available tools

list_tasks

List all tasks with optional status_filter and show results in the MCP client or via API.

create_task

Create a new task with title, description, optional priority and assignee.

update_task_status

Update the status of an existing task by task_id.

get_task_statistics

Return comprehensive task statistics and analytics.

search_tasks

Search tasks by title or description.

export_tasks_to_csv

Export all tasks to a CSV file for offline analysis.

gemini

Interact with Google Gemini for AI-powered prompts and responses.