Connect AI models and tools to SurrealDB through the Model Context Protocol. No wrapper APIs or custom integration code.
Access all your data models
Query documents, graphs, vectors, and relational data through a unified interface. Your AI can access any data type with a single query language.
Real-time AI interactions
AI models query live data, traverse graphs, and access vector embeddings - all through a single MCP connection.
Secure AI data access
Built-in authentication and row-level permissions ensure AI tools only access authorised data.
Developer-friendly integrations
Install the VSCode or Cursor extension, point it at your SurrealDB instance, and start querying.
Native vector search
Run semantic search, similarity matching, and RAG pipelines using SurrealDB's built-in vector indexes.
Graph-powered AI
Traverse relationships and analyse multi-hop connections through SurrealDB's native graph engine.
One query language for AI
SurrealQL queries documents, graphs, vectors, and time-series in one statement. No separate data stores to connect.
MCP across the platform
Every product in the SurrealDB platform exposes MCP tool calls. Connect your AI tools to the database, agent memory, or cloud - through a single protocol.
Default stdio transport for local development and IDE integration. Simple, efficient communication with no network overhead.
Environment adaptability
Choose the right transport for your deployment: stdio for local tools, HTTP with SSE for remote agents.
Cross-platform compatibility
Runs on Windows, macOS, and Linux with identical behaviour across all three.
Run locally from your IDE
The MCP server runs locally via stdio. Connect to SurrealDB instances from any IDE, framework, or AI tool. Start an in-memory or local database directly from your AI tool.