Modelwatch

Overview

Modelwatch is a Model Context Protocol server that provides continuous behavioral drift monitoring for LLM applications. It detects silent provider model updates that could affect your app's performance and reliability without your knowledge.

The server runs locally through the npm package modelwatch-mcp, which communicates via stdio. You'll need to set the MODELWATCH_API_KEY environment variable to authenticate with the service. The latest release is v0.1.1, and the source code is available at bch1212/modelwatch on GitHub.

Modelwatch is listed in the AI & Agents category on mcp.site and works with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and Zed. If you maintain this server, you can claim this listing to verify ownership and keep information up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "modelwatch": {
      "command": "npx",
      "args": [
        "-y",
        "modelwatch-mcp"
      ],
      "env": {
        "MODELWATCH_API_KEY": "<modelwatch-api-key>"
      }
    }
  }
}

Requires MODELWATCH_API_KEY