How coleam00/supabase-mcp works
Project type and purpose
• This is a lightweight microservice (MCP server) packaged as a Docker container.
• Purpose: Expose CRUD tools (read, create, update, delete) over the Model Context Protocol (MCP) so that AI assistants can perform operations against a Supabase database through a standardized stdio interface.
File structure
• Top level:
• Dockerfile – builds the container image “mcp/supabase.”
• example_mcp_config.json – shows how an AI assistant configures this server.
• requirements.txt – Python dependencies.
• supabase_mcp/ – Python package containing:
– server.py (main entry point implementing the MCP interface)
– tests/ (unit tests for server logic)
• Supporting docs (README, PLANNING.md, TASKS.md) aid development but are outside the runtime path.
README for additional insights
• Confirms the server listens on stdio, implements the MCP spec, and expects environment variables SUPABASE_URL and SUPABASE_SERVICE_KEY.
• Describes the four tools (read_table_rows, create_table_records, update_table_records, delete_table_records).
• Shows workflow: AI assistant ⟷ MCP server (stdio) ⟷ Supabase REST/JS SDK ⟷ Supabase database.