MCP (Model Context Protocol) Server

What is MCP (Model Context Protocol) Server?

A Python-based server that integrates PostgreSQL databases with AI providers (Anthropic, Google) using the Model Context Protocol, supporting flexible communication transports (SSE or stdio).

How to use MCP (Model Context Protocol) Server?

Clone the repository, create a Python virtual environment, install dependencies, configure environment by copying .env.example to .env.dev and filling in database and API credentials, then run python server.py for the server and python client.py for client interaction.

Key features of MCP (Model Context Protocol) Server

• Secure configuration management
• PostgreSQL database integration
• Multi-provider AI model support
• Flexible communication transport
• Extensible tool registration

Use cases of MCP (Model Context Protocol) Server

• Retrieve database table listings via AI prompts
• Extend server with custom tool functions using decorators
• Interact with PostgreSQL through a client script

FAQ from MCP (Model Context Protocol) Server

What are the prerequisites?

Python 3.10+, a PostgreSQL instance, and API access to Anthropic and/or Google AI providers.

How do I install and configure the server?

Clone the repo, create a virtual environment, run pip install -r requirements.txt, copy .env.example to .env.dev, and fill in database credentials, API keys, and transport settings.

Which AI providers are supported?

Anthropic (Claude models) and Google (Gemini models). The default provider is set via the USE_PROVIDER config.

What transport options are available?

The server supports both SSE (/sse endpoint) and stdio transports, configured via the TRANSPORT environment variable.

How can I add a new tool?

Use the @app.tool() decorator above a function definition; the function will be registered as a custom tool.