Mcp Scorecard

Overview

Mcp Scorecard evaluates the agent-readiness of any MCP server through a standardized assessment. It runs 10 checks to produce a 0-100 score along with actionable findings that help you understand where your MCP implementation stands and what needs improvement.

The server runs locally via npx -y mcp-scorecard with no authentication required. It speaks the Model Context Protocol over stdio, making it immediately available to use. The latest release is v0.5.1, distributed under the MIT license with full source code at davidmosiah/mcp-scorecard on GitHub.

Mcp Scorecard is listed in the AI & Agents category and integrates with Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and other MCP-compatible clients. If you maintain this server, claim your listing to earn a Verified badge and ensure the information stays accurate.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "mcp_scorecard": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-scorecard"
      ]
    }
  }
}