This project enables natural language querying of PostgreSQL databases through a Model Context Protocol (MCP) server and Python client. It lets LLMs discover schema, describe tables, and execute safe, read-only SQL via conversational tools. Data onboarding and schema guidance are managed by clear, domain-specific system prompts. With Docker-based deployment, you can quickly connect, seed, and chat with your data—making database access as simple as a conversation. For setup and advanced usage, please read the README files in both the mcp-server and mcp-client directories.
- Connect the backend to a chat client app for a complete conversational interface.
- Add support for connecting to different MCP servers, enabling multi-database or multi-backend querying.