Overview
What is mcp-server-weaviate?
mcp-server-weaviate is an MCP (Model Context Protocol) server that connects AI agents to Weaviate, a vector database. It enables tools for searching and storing vector data, using OpenAI embeddings, and is designed for developers building AI-powered applications with vector search.
How to use mcp-server-weaviate?
Install via Smithery with npx -y @smithery/cli install @weaviate/mcp-server-weaviate --client claude, or manually configure Claude Desktop by adding a JSON entry to claude_desktop_config.json. The configuration requires --weaviate-url, --weaviate-api-key, --search-collection-name, --store-collection-name, and --openai-api-key arguments.
Key features of mcp-server-weaviate
- Connects to Weaviate vector database for AI agents
- Provides search and store tools for vector data
- Uses OpenAI embeddings for vector operations
- Supports configurable search and store collections
- Easy installation via Smithery or manual configuration
Use cases of mcp-server-weaviate
- Semantic search over large document collections
- Storing and retrieving vector embeddings for LLM context
- Building AI assistants that query Weaviate databases
- Enabling vector-based retrieval for RAG pipelines
FAQ from mcp-server-weaviate
How do I install mcp-server-weaviate?
Install via Smithery using npx -y @smithery/cli install @weaviate/mcp-server-weaviate --client claude, or manually configure Claude Desktop with the provided JSON config.
What are the prerequisites for using mcp-server-weaviate?
You need to have uv installed, clone the repository, and have a Weaviate instance with an API key. An OpenAI API key is also required for embeddings.
What configuration parameters are required?
The server requires --weaviate-url, --weaviate-api-key, --search-collection-name, --store-collection-name, and --openai-api-key. These are passed as command-line arguments.
Do I need separate collections for search and store?
Yes, the configuration expects separate collection names for search and store operations, specified via --search-collection-name and --store-collection-name.
How does mcp-server-weaviate use OpenAI?
The server uses an OpenAI API key to generate vector embeddings for both search and store operations, enabling semantic search capabilities.
