Model Context Protocol (MCP)

​

Overview

Morphik provides native support for the Model Context Protocol (MCP) - an open standard that allows AI models like Claude to directly access structured data and knowledge bases.

With Morphik’s MCP integration, you can:

• Let Claude, or other MCP supported models to search through your documents based on natural language queries
• Retrieve knowledge from your Morphik database without writing complex code
• Give AI models direct access to your organization’s information
• Enable more accurate, context-aware responses using your own data

​

Setup

Morphik’s MCP support is provided through the @morphik/mcp package, which acts as a bridge between Morphik and MCP-compatible AI assistants.

​

Prerequisites

• A running Morphik server (local or cloud)
• Node.js 16.0.0 or higher

​

Configuration with Claude Desktop

To integrate Morphik with Claude Desktop:

1. Open your Claude Desktop configuration file:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

2. Add the Morphik MCP configuration:

{
  "mcpServers": {
    "morphik": {
      "command": "npx",
      "args": [
        "-y",
        "@morphik/mcp",
        "--uri=your-morphik-server-uri"
      ]
    }
  }
}

For local Morphik server, you can either omit the --uri parameter or use --uri=local:

{
  "mcpServers": {
    "morphik": {
      "command": "npx",
      "args": [
        "-y",
        "@morphik/mcp"
      ]
    }
  }
}

​

Using with Claude

Once configured:

1. Close and restart Claude Desktop.
2. Claude will now have access to the following tools:
   • retrieve-chunks: Search for relevant chunks from your knowledge base
   • retrieve-docs: Search for entire documents in your knowledge base
   • query: Generate a completion using relevant context
   • list-documents: List available documents
   • get-document: Get a specific document’s details
   • ingest-text: Add new text to your knowledge base
   • delete-document: Remove a document from your knowledge base