Mixpeek Logo
    Start free
    ai ml
    Featured

    MCP Server

    Give AI agents perception over multimodal data via the Model Context Protocol

    A Model Context Protocol (MCP) server that exposes Mixpeek's multimodal search, classification, and metadata extraction as agent-callable tools. Any MCP-compatible AI agent (Claude, ChatGPT, Cursor, Windsurf, Cline) can discover and invoke these tools to search video, classify images, extract metadata from documents, and detect brands or faces -- all through natural language.

    Built for AI agent developers, platform teams building agent toolkits, and enterprises adding multimodal perception to their AI workflows.

    MCP
    Model Context Protocol
    AI agents
    tool use
    multimodal search
    video search
    Quick Install:
    pip install mixpeek-mcp

    Use Cases

    1

    Give Claude, ChatGPT, or custom agents the ability to search your video archive

    2

    Enable AI agents to classify content against brand safety or IAB taxonomies

    3

    Build compliance review agents that reason across documents, images, and recordings

    4

    Add visual product search to customer support agents

    5

    Create media intelligence agents that monitor and alert on content signals

    Features

    Semantic search across video, images, audio, and documents via natural language
    IAB v3.0 taxonomy classification for any media file
    Brand safety and logo detection tools
    Face detection and recognition tools
    Structured metadata extraction (objects, scenes, text, audio)
    Namespace-scoped queries for multi-tenant environments
    stdio and HTTP+SSE transport support
    Confidence scores on all results for agent decision-making

    Example

    Claude Desktop config (mcp.json)

    {
      "mcpServers": {
        "mixpeek": {
          "command": "python",
          "args": ["-m", "mixpeek_mcp"],
          "env": {
            "MIXPEEK_API_KEY": "your_api_key",
            "MIXPEEK_NAMESPACE": "your_namespace"
          }
        }
      }
    }

    Agent invocation example

    Agent: "Find all video segments showing a red Nike logo"
    

    Tool call: search
      query: "red Nike logo"
      modalities: ["video", "image"]
    

    Result:
      3 matches found (confidence: 0.91-0.96)
      warehouse-footage.mp4 @ 02:34-02:41
      product-shoot.mp4 @ 00:12-00:18
      storefront-cam.jpg (full frame)

    Get Started

    Integrations

    Claude Desktop
    Claude Code
    Cursor
    Windsurf
    Cline
    OpenAI Assistants
    LangChain
    CrewAI

    Details

    LicenseApache 2.0
    Categoryai ml
    Registrypypi
    Latency50-500ms per tool call (varies by modality)
    Deploymentpip install / Docker / remote HTTP+SSE
    Typical UseAgent-driven media search, classification, and monitoring

    Resources

    Documentation

    Learn how to use this connector

    Quick Info

    LicenseApache 2.0
    Categoryai ml
    Registrypypi
    Latency50-500ms per tool call (varies by modality)
    Deploymentpip install / Docker / remote HTTP+SSE
    Typical UseAgent-driven media search, classification, and monitoring

    Frequently Asked Questions

    Ready to integrate?

    Get started with MCP Server in minutes. Check out the documentation or explore the source code on GitHub.