{
  "name": "io.github.chroma-core/chroma-mcp",
  "description": "The official Chroma MCP connects an AI agent to a Chroma collection — the in-process vector database used as the default in LangChain, LlamaIndex, and many agent stacks. Add documents, run semantic queries, manage collections, and inspect embeddings. Works against an in-process Chroma, a persistent local path, or hosted Chroma Cloud.",
  "status": "active",
  "version": "latest",
  "_meta": {
    "com.top-mcps/lastVerified": "2026-06-02",
    "com.top-mcps/score": {
      "value": 72,
      "rubricVersion": "2026-04",
      "lastComputed": "2026-06-02T13:16:41.658Z",
      "methodology": "https://top-mcps.com/about/methodology"
    },
    "com.top-mcps/stats": {
      "fetchedAt": "2026-06-02T13:16:41.453Z",
      "githubStars": 554,
      "lastCommitAt": "2025-09-17T20:20:13Z"
    }
  },
  "repository": {
    "url": "https://github.com/chroma-core/chroma-mcp",
    "source": "github"
  },
  "author": {
    "name": "Chroma"
  },
  "packages": [
    {
      "registryType": "pypi",
      "registryBaseUrl": "https://pypi.org",
      "identifier": "chroma-mcp",
      "version": "latest",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "name": "CHROMA_PATH",
          "description": "CHROMA_PATH — required credential for Chroma.",
          "isRequired": true,
          "isSecret": true
        },
        {
          "name": "CHROMA_HOST",
          "description": "CHROMA_HOST — required credential for Chroma.",
          "isRequired": true,
          "isSecret": true
        },
        {
          "name": "CHROMA_API_KEY",
          "description": "CHROMA_API_KEY — required credential for Chroma.",
          "isRequired": true,
          "isSecret": true
        }
      ]
    }
  ],
  "capabilities": {
    "tools": [
      {
        "name": "list_collections",
        "description": "List collections in the Chroma instance.",
        "sideEffect": "read"
      },
      {
        "name": "create_collection",
        "description": "Create a new collection.",
        "sideEffect": "write",
        "args": [
          {
            "name": "name",
            "type": "string",
            "required": true
          }
        ]
      },
      {
        "name": "add_documents",
        "description": "Add documents to a collection.",
        "sideEffect": "write",
        "args": [
          {
            "name": "collection",
            "type": "string",
            "required": true
          },
          {
            "name": "documents",
            "type": "array",
            "required": true
          },
          {
            "name": "metadatas",
            "type": "array",
            "required": false
          }
        ]
      },
      {
        "name": "query",
        "description": "Run a semantic query with optional metadata filters.",
        "sideEffect": "read",
        "args": [
          {
            "name": "collection",
            "type": "string",
            "required": true
          },
          {
            "name": "query_text",
            "type": "string",
            "required": true
          },
          {
            "name": "n_results",
            "type": "number",
            "required": false
          }
        ]
      }
    ],
    "security": {
      "scope": "read-write",
      "sandbox": "Local mode runs entirely in the MCP process — data lives at CHROMA_PATH. Cloud mode authenticates with an API key scoped to one tenant. No outbound network calls in local mode.",
      "gotchas": [
        "Local mode writes to disk on every add — back up the CHROMA_PATH directory if it matters.",
        "Cloud API keys are full-tenant — scope by tenant rather than relying on collection-level isolation.",
        "The default embedding function downloads model weights on first run — pre-warm in CI to avoid first-call latency."
      ]
    }
  },
  "_links": {
    "html": "https://top-mcps.com/mcp/chroma",
    "markdown": "https://top-mcps.com/mcp/chroma.md",
    "methodology": "https://top-mcps.com/about/methodology"
  }
}