isagentready-mcp

A Model Context Protocol (MCP) server for the IsAgentReady API. Scan any website for AI agent readiness and get detailed reports with scores, letter grades, and actionable recommendations across 5 categories: Discovery, Structured Data, Semantics, Agent Protocols, and Security.

Add To Your Editor

CLI one-liners — the fastest way to get started. Pick your tool:

Claude Code

claude mcp add isagentready-mcp -- npx -y isagentready-mcp

Codex CLI (OpenAI)

codex mcp add isagentready-mcp -- npx -y isagentready-mcp

Gemini CLI (Google)

gemini mcp add isagentready-mcp -- npx -y isagentready-mcp

VS Code (Copilot)

Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P) > MCP: Add Server > select Command (stdio).

Or add to .vscode/mcp.json in your project directory:

{
  "servers": {
    "isagentready-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "isagentready-mcp"]
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project directory (or ~/.cursor/mcp.json for global):

{
  "mcpServers": {
    "isagentready-mcp": {
      "command": "npx",
      "args": ["-y", "isagentready-mcp"]
    }
  }
}

Supported Clients

This MCP server works with any client that supports the Model Context Protocol, including:

{
  "mcpServers": {
    "isagentready-mcp": {
      "command": "npx",
      "args": ["-y", "isagentready-mcp"]
    }
  }
}

{
  "mcpServers": {
    "isagentready-mcp": {
      "command": "npx",
      "args": ["-y", "isagentready-mcp"]
    }
  }
}

{
  "context_servers": {
    "isagentready-mcp": {
      "command": "npx",
      "args": ["-y", "isagentready-mcp"]
    }
  }
}

docker run -i --rm ghcr.io/bartwaardenburg/isagentready-mcp

[mcp_servers.isagentready-mcp]
command = "npx"
args = ["-y", "isagentready-mcp"]

Features

• 3 MCP tools for scanning, retrieving results, and browsing rankings
• No API key required — public API, zero configuration needed
• Built-in caching with configurable TTL and automatic invalidation on scans
• Retry logic with exponential backoff and Retry-After header support
• Structured content returned alongside human-readable text
• Toolset filtering to expose only the tool categories you need
• Actionable error messages with context-aware recovery suggestions
• Docker support for containerized deployment
• 97 unit tests across 7 test files

Configuration

All configuration is optional via environment variables:

Example with configuration

{
  "mcpServers": {
    "isagentready-mcp": {
      "command": "npx",
      "args": ["-y", "isagentready-mcp"],
      "env": {
        "ISAGENTREADY_CACHE_TTL": "300",
        "ISAGENTREADY_MAX_RETRIES": "5"
      }
    }
  }
}

Available Tools

Scans

Rankings

Toolset Filtering

Reduce context window usage by enabling only the tool categories you need. Set the ISAGENTREADY_TOOLSETS environment variable to a comma-separated list:

ISAGENTREADY_TOOLSETS=scans

When not set, all toolsets are enabled. Invalid names are ignored; if all names are invalid, all toolsets are enabled as a fallback.

Security Notes

• Trust model: Any prompt or agent allowed to call this MCP server can trigger website scans and retrieve results via the IsAgentReady API.
• No credentials required: This server uses a public API, so there are no secrets to protect. However, scan requests reveal which domains you are evaluating.
• Team config governance: Keep shared MCP config in version control and require review for changes to command/args/env/toolset filtering.

Example Usage

Once connected, you can interact with the IsAgentReady API using natural language:

• "Scan example.com for AI agent readiness"
• "Check if the scan results for example.com are ready"
• "Show me the top AI-ready websites"
• "Search the rankings for sites in the high grade range"
• "How does my site compare to others?"

Scan + fix workflow

Combine with the isagentready-skills agent skills for a full scan, fix, and re-verify workflow:

> Scan example.com for AI agent readiness

Scan enqueued for example.com
Status: pending

> Check the results for example.com

Domain: example.com
Grade: B (72/100)
Status: completed

Discovery — 15/20 (75%, weight: 30%)
  [PASS] 1.1 robots.txt (5/5)
  [FAIL] 1.2 sitemap.xml (0/5)
    Recommendation: Add a sitemap.xml file...

> Fix the failing checkpoints

Community

• Support: SUPPORT.md
• Security reporting: SECURITY.md
• Contributing guidelines: CONTRIBUTING.md
• Bug reports and feature requests: Issues

Development

# Install dependencies
pnpm install

# Run in development mode
pnpm dev

# Build for production
pnpm build

# Run tests
pnpm test

# Type check
pnpm typecheck

Project Structure

src/
  index.ts              # Entry point (stdio transport)
  server.ts             # MCP server setup, toolset filtering
  client.ts             # IsAgentReady API HTTP client with caching and retry
  cache.ts              # TTL-based in-memory response cache
  types.ts              # TypeScript interfaces
  format.ts             # Output formatting helpers
  tool-result.ts        # Error formatting with recovery suggestions
  update-checker.ts     # NPM update notifications
  tools/
    scans.ts            # scan_website, get_scan_results
    rankings.ts         # get_rankings

Requirements

• Node.js >= 20

Related

• isagentready-skills — Agent skills for fixing AI readiness issues identified by scans
• IsAgentReady.com — The web scanner

License

MIT — see LICENSE for details.