firebase-ml-api
0
The MCP Server project facilitates multi-agent conversation protocols and is designed to work with a specified OpenAPI URL. It includes tools for linting, testing, and publishing, supporting various transport modes and configuration through environment variables.
MCP Server
This project is an MCP (Multi-Agent Conversation Protocol) Server for a specific OpenAPI URL. It is auto-generated using AG2's MCP builder.
Prerequisites
- Python 3.9+
- pip and uv
Installation
- Clone the repository and navigate into the directory.
- Install dependencies through the devcontainer setup script or manually with pip. An alternative is to use uv for installation.
Development
- Utilizes ruff for linting, mypy for static type checking, and pytest for testing.
- Static analysis includes tools like mypy, bandit, semgrep.
Running the Server
- Can be started using the main.py script with different transport modes (e.g., stdio, sse).
- Configured using environment variables.
Building and Publishing
- Uses Hatch for building and publishing processes.