# AgentLighthouse — Full Documentation for AI Agents
> AI & Agent Readiness Scanner by PHRIXUS

AgentLighthouse is an open-source tool that evaluates websites for AI and Agent Readiness, checking structured data, semantic HTML, AI crawler compatibility, security headers, and performance metrics.

> Core Audit Categories

### Security (18pts)
- SSL certificate validation and expiry tracking
- Security headers: HSTS, X-Frame-Options, X-Content-Type-Options, CSP, Referrer-Policy, Permissions-Policy

### SEO & Metadata (8pts)
- Title tag, meta description, Open Graph, Twitter Card
- Canonical URL, robots.txt, sitemap.xml

### AI & Agent Readiness (21pts)
- llms.txt and llms-full.txt detection and quality scoring
- Markdown for Agents (Accept header negotiation)
- Semantic HTML tag usage (header, nav, main, footer, section, article)
- Heading structure hierarchy (H1-H6)
- X-Robots-Tag and AI bot access verification

### Structured Data (14pts)
- JSON-LD presence and parse validation
- Schema.org type coverage (WebApplication, Organization, SoftwareSourceCode, etc.)

> API Reference

### POST /api/audit
Request: {"url": "<target-url>"}
Response: {score, raw_score, max_score, level, checks[], url}

### GET /health
Response: {"status": "ok", "version": "0.2.0"}

> Scoring System

Each check has a max_score. Total raw_score / max_score normalized to 0-100.
Status levels: excellent (90+), good (70+), fair (50+), poor (<50)

> Integration

AgentLighthouse can be integrated into CI/CD pipelines for continuous AI readiness monitoring.\nPOST /api/audit accepts any public URL and returns machine-readable JSON.\n\n## Project\n- Repository: https://github.com/phrixus-ai/agent-lighthouse\n- License: MIT
- Author: PHRIXUS