ChromaDB MCP Server

About

## ChromaDB MCP Server: Open Source Vector Database The **ChromaDB MCP Server** integrates the leading open-source vector database into Google Antigravity. Chroma makes it simple to build AI applications with embeddings, providing semantic search and retrieval capabilities essential for RAG systems. ### Why ChromaDB MCP? ChromaDB simplifies vector storage: - **Developer Friendly**: Simple, intuitive API - **Embedded Mode**: Run in-process without servers - **Persistent Storage**: Save collections to disk - **Multi-Modal**: Store embeddings with metadata - **Open Source**: Full control over your data ### Key Features #### 1. Collection Management ```python import chromadb client = chromadb.Client() # Create collection with embedding function collection = client.create_collection( name="documents", metadata={"hnsw:space": "cosine"} ) # Add documents with auto-embedding collection.add( documents=["Machine learning is fascinating", "AI transforms industries"], metadatas=[{"source": "blog"}, {"source": "article"}], ids=["doc1", "doc2"] ) ``` #### 2. Semantic Search ```python # Query with natural language results = collection.query( query_texts=["How does ML work?"], n_results=5, include=["documents", "distances", "metadatas"] ) for doc, distance in zip(results["documents"][0], results["distances"][0]): print(f"{distance:.3f}: {doc}") ``` #### 3. Filtering ```python # Combine semantic search with metadata filters results = collection.query( query_texts=["AI applications"], n_results=10, where={"source": "article"}, where_document={"$contains": "machine learning"} ) ``` ### Configuration ```json { "mcpServers": { "chromadb": { "command": "npx", "args": ["-y", "@anthropic/mcp-chromadb"], "env": { "CHROMA_HOST": "localhost", "CHROMA_PORT": "8000" } } } } ``` ### Use Cases **RAG Applications**: Store and retrieve document chunks for question-answering systems. **Semantic Search**: Build search that understands meaning, not just keywords. **Recommendation**: Find similar items based on content embeddings. The ChromaDB MCP Server enables semantic search and retrieval in Antigravity.

Installation

{
  "mcpServers": {
    "chromadb": {
      "mcpServers": {
        "chromadb": {
          "env": {
            "CHROMA_HOST": "localhost",
            "CHROMA_PORT": "8000"
          },
          "args": [
            "chromadb-mcp"
          ],
          "command": "uvx"
        }
      }
    }
  }
}

How to Use

1. 11. Set ChromaDB server host and port
2. 22. Supports semantic document search
3. 33. Can index codebases locally