The Weaviate MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Weaviate account. It provides structured and secure access to your Weaviate Cloud collections, so your agent can answer natural-language questions, run AI-powered searches, apply structured filters, perform aggregations, and return matching data on your behalf.
- Natural-language questions: Have your agent ask questions about your Weaviate collections and turn them into searches or aggregations before providing a final answer.
- Semantic collection search: Let the agent find relevant collection data using AI-powered semantic search rather than relying only on exact keyword matches.
- Structured filtering: Direct your agent to combine semantic search with structured filters to narrow results to the data that matches your criteria.
- Data aggregation: Instruct your agent to generate aggregations from your collection data when answering questions that require grouped or calculated results.
- Matching result retrieval: Have your agent return the Weaviate objects that match a search, so you can work directly with the relevant collection data.