ChromaDB MCP Server

Connects AI agents to ChromaDB vector databases for semantic search and retrieval augmented generation (RAG) workflows. Supports creating collections, upserting documents with embeddings, querying by semantic similarity, and managing metadata filters. Ideal for knowledge base and document retrieval applications.

{
  "mcpServers": {
    "chromadb": {
      "command": "uvx",
      "args": [
        "chroma-mcp-server@latest"
      ],
      "env": {
        "CHROMA_HOST": "<YOUR_CHROMA_HOST>",
        "CHROMA_PORT": "<YOUR_CHROMA_PORT>",
        "CHROMA_AUTH_TOKEN": "<YOUR_CHROMA_AUTH_TOKEN>"
      }
    }
  }
}

{
  "type": "object",
  "required": [
    "name"
  ],
  "properties": {
    "name": {
      "type": "string",
      "description": "Collection name"
    },
    "metadata": {
      "type": "object",
      "description": "Optional collection metadata"
    },
    "embedding_function": {
      "type": "string",
      "description": "Embedding function to use (openai, cohere, sentence-transformers)"
    }
  }
}

{
  "type": "object",
  "required": [
    "collection",
    "documents",
    "ids"
  ],
  "properties": {
    "ids": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Unique IDs for each document"
    },
    "documents": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Text documents to embed and store"
    },
    "metadatas": {
      "type": "array",
      "items": {
        "type": "object"
      },
      "description": "Metadata for each document"
    },
    "collection": {
      "type": "string",
      "description": "Collection name"
    }
  }
}

{
  "type": "object",
  "required": [
    "collection",
    "query_text"
  ],
  "properties": {
    "where": {
      "type": "object",
      "description": "Metadata filter conditions"
    },
    "n_results": {
      "type": "integer",
      "description": "Number of results to return"
    },
    "collection": {
      "type": "string",
      "description": "Collection name"
    },
    "query_text": {
      "type": "string",
      "description": "Text to search for"
    }
  }
}

{
  "type": "object",
  "properties": {}
}