home / mcp / mcp metaso mcp server

MCP Metaso MCP Server

Provides multi-source MCP search across web, documents, scholars, images, videos, and podcasts with content extraction and MCP protocol compatibility.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "hundunonline-mcp-metaso": {
      "url": "https://mcp.metaso.example/mcp",
      "headers": {
        "METASO_API_KEY": "your-api-key-here"
      }
    }
  }
}

MCP Metaso is a Python-based MCP server that enables a versatile search experience across web, documents, scholarly works, images, videos, and podcasts. It processes pages to extract content and exposes an MCP-compatible interface so you can plug it into clients like Claude Desktop for fast, type-safe searching across multiple sources.

How to use

You run the MCP Metaso server locally and connect your MCP client to it. The server supports multiple search scopes, including webpage, document, scholar, image, video, and podcast. You can start a local instance and use your preferred MCP client to send search queries and receive structured results.

How to install

Prerequisites: ensure you have Python 3.10 or newer installed on your system.

Install dependencies from the requirements file.

Set your API key for Metaso access.

Start the server and verify it is responding to test queries.

Claude Desktop integration

There is built-in support to integrate the MCP Metaso server with Claude Desktop using an extension format. You can install the extension, configure the API key, and enable the server so Claude Desktop can query it directly.

For a quick setup, you can configure Claude Desktop to launch the local MCP server via the provided startup script and environment variable for METASO_API_KEY.

You can also configure Claude Desktop manually by editing the client configuration to point at your local server and including METASO_API_KEY in the environment.

Configuration and troubleshooting

If the server doesn’t start or respond, check that the API key is set, confirm the server path used to start the process, and look at the application logs for error messages.

Common steps to verify startup include running the start command, then sending a simple test query to confirm a valid response.

Available tools

metaso_search

Multi-dimension search across six scopes: webpage, document, scholar, image, video, and podcast.

metaso_reader

Web page content extraction tool that outputs in markdown or json.