Announcing the Supabase Remote MCP Server

Supabase has enhanced its Model Context Protocol (MCP), a standard for connecting Large Language Models (LLMs) to platforms like Supabase, by adding new features and improving existing ones. Initially launched in April, the MCP server facilitates the integration of AI tools such as Cursor and Claude Code with Supabase, enabling users to connect via a single URL. The recent updates include the addition of Remote HTTP transport, feature groups for tool selection, a document search tool powered by a GraphQL-based API, and security and performance advisors to help AI agents follow best practices. Additionally, Supabase has integrated storage capabilities, allowing AI agents to manage project buckets. Despite the knowledge cutoff challenges inherent in LLMs, these updates aim to enhance productivity and provide AI-first user experiences while maintaining security and performance standards.