Agent Lighthouse MCP Server
Enables AI coding agents like Claude and Cursor to audit websites for AI agent readiness, checking 199 rules across agentic discovery, content structure, and technical SEO.
README
<div align="center"> <h1>๐ผ Agent Lighthouse</h1> <p><strong>Lighthouse, but for AI agents.</strong></p> <p>Audit whether ChatGPT, Claude, Perplexity, MCP clients, AI crawlers, and agentic browsers can discover, parse, cite, and act on your website.</p> <p> <a href="https://www.npmjs.com/package/@forkpoint/agent-lighthouse"><img alt="npm version" src="https://img.shields.io/npm/v/@forkpoint/agent-lighthouse.svg?color=4f46e5"></a> <a href="https://www.npmjs.com/package/@forkpoint/agent-lighthouse"><img alt="npm downloads" src="https://img.shields.io/npm/dm/@forkpoint/agent-lighthouse.svg"></a> <a href="https://github.com/ForkPoint/agent-lighthouse/actions/workflows/ci.yml"><img alt="CI" src="https://github.com/ForkPoint/agent-lighthouse/actions/workflows/ci.yml/badge.svg"></a> <a href="https://forkpoint.github.io/agent-lighthouse/"><img alt="Docs" src="https://img.shields.io/badge/docs-GitHub%20Pages-4f46e5"></a> <a href="./LICENSE"><img alt="License: Apache-2.0" src="https://img.shields.io/badge/license-Apache--2.0-blue"></a> </p> <p> <a href="https://forkpoint.github.io/agent-lighthouse/">Documentation</a> ยท <a href="https://www.npmjs.com/package/@forkpoint/agent-lighthouse">npm</a> ยท <a href="https://github.com/ForkPoint/agent-lighthouse/issues/new?template=site-score.yml">Share a site score</a> </p> </div>
โก Quickstart
Run a zero-install scan directly in your terminal. The --view flag opens the standalone HTML report for screenshots, stakeholder review, and pull-request artifacts.
# Instant audit (prints terminal report & generates HTML + JSON reports)
npx @forkpoint/agent-lighthouse https://yourstore.com
# Open the standalone HTML report in your browser
npx @forkpoint/agent-lighthouse https://yourstore.com --view
# Run in CI and fail if score is below threshold
npx @forkpoint/agent-lighthouse https://staging.yourstore.com --min-score 85
Agent Lighthouse checks 207 rules covering llms.txt, robots.txt crawler policy, Schema.org, OpenAPI discovery, WebMCP action surfaces, AEO/GEO content structure, accessibility, and technical readiness.
๐ฏ What Agent Lighthouse Checks
Agent Lighthouse evaluates websites across 10 audit categories grouped into 3 readiness pillars:
โโโ 1. Agentic Readiness
โ โโโ AI Agent Tools & Action Surfaces (WebMCP manifests, OpenAPI specs, agents.json, ai-plugin.json)
โ โโโ Content Discoverability (llms.txt, llms-full.txt, sitemaps, commerce links)
โ โโโ AI Crawler Permissions (robots.txt rules for GPTBot, ClaudeBot, PerplexityBot, etc.)
โ
โโโ 2. AI Search Optimization
โ โโโ Answer Engine Optimization (AEO) (direct answerability, step lists, table schemas)
โ โโโ Generative Engine Optimization (GEO) (unique data density, authoritative citations)
โ
โโโ 3. Technical Foundation
โโโ Structured Data & Schema Markup (Schema.org Product, Offer, SKU, GTIN, Organization)
โโโ Meta Tags & AI Head Elements (AI content declarations, canonicals, Open Graph)
โโโ Semantic HTML & Content Structure (Headings hierarchy, landmarks, semantic tags)
โโโ Accessibility & Agent Interaction (Form labels, button roles, interactable elements)
โโโ Technical Readiness & Security (HTTPS, security.txt, TTFB response latency)
๐ฆ Packages & Architecture
This repository is organized as a lightweight pnpm monorepo published under the @forkpoint scope:
| Package | npm Package | Description |
|---|---|---|
packages/cli |
@forkpoint/agent-lighthouse |
Main CLI binary (npx @forkpoint/agent-lighthouse <url>). |
packages/core |
@forkpoint/agent-lighthouse-core |
Core gatherer-audit engine, scoring algorithms, and types. |
packages/report |
@forkpoint/agent-lighthouse-report |
Standalone HTML, Markdown, and unified report view-model. |
packages/mcp |
@forkpoint/agent-lighthouse-mcp |
Model Context Protocol (MCP) server for Claude / Cursor / IDEs. |
๐ก๏ธ GitHub Actions CI
Use Agent Lighthouse as a pull-request gate for agentic readiness regressions:
name: Agent Lighthouse
on:
pull_request:
branches: [main]
jobs:
agent-lighthouse:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: ForkPoint/agent-lighthouse@main
with:
url: https://staging.yourstore.com
preset: ecommerce
min-score: "85"
github-token: ${{ secrets.GITHUB_TOKEN }}
The action generates terminal, HTML, JSON, and Markdown reports. Set comment-on-pr: true with github-token to post the Markdown summary on pull requests. See the marketplace setup guide for release-ready examples.
๐ป Programmatic Node.js / TypeScript SDK
import { runScan } from "@forkpoint/agent-lighthouse-core";
import {
buildReportView,
generateHtmlReport,
} from "@forkpoint/agent-lighthouse-report";
const report = await runScan("https://example.com");
const view = buildReportView(report);
console.log(`Overall Score: ${view.overallScore}/100 (${view.scoreTier})`);
// Generate standalone HTML report
const html = generateHtmlReport(report);
๐ค Model Context Protocol (MCP) Server
Add Agent Lighthouse to your Claude Desktop or Cursor IDE to let AI coding agents audit live staging URLs:
{
"mcpServers": {
"agent-lighthouse": {
"command": "npx",
"args": ["-y", "@forkpoint/agent-lighthouse-mcp"]
}
}
}
๐ฃ Share Your Score
Generated reports are standalone files, so teams can attach them to pull requests, send them to clients, or publish before/after improvements.
[](https://github.com/ForkPoint/agent-lighthouse)
If you run Agent Lighthouse on a public site, share the result through the site score template. Good examples help other developers learn what agent-ready sites look like.
More launch material lives in:
- Promotion kit
- Benchmark report
- Badge generator notes
- Launch post drafts
- Outreach templates
- Terminal demo transcript
- Docs homepage screenshot
- Generated report screenshot
- Badge generator screenshot
- Report preview asset
- MCP setup screenshot asset
๐ ๏ธ Development
# Clone the repository
git clone https://github.com/ForkPoint/agent-lighthouse.git
cd agent-lighthouse
# Install dependencies
pnpm install
# Build all packages
pnpm build
# Run unit tests
pnpm test
๐ License
Apache-2.0 ยฉ ForkPoint
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.