AgentCheck

AI is recommending your competitors. Not you.

Free AI visibility scanner — check if ChatGPT, Claude, Perplexity, Gemini, Copilot & Google AI Overview can find your website.

Live Scanner · Report a Bug · Request Feature

The next generation of customers won't Google you — they'll ask AI. AgentCheck scans any website and shows you exactly which AI agents can find you, which can't, and what to fix.

Your GEO & LLM SEO toolkit. Free, open source.

The Problem

AI agents are reshaping discovery. Millions of people skip Google and ask ChatGPT, Claude, or Perplexity for recommendations. If your website isn't machine-readable, you're invisible.

No llms.txt or ai.txt — AI agents can't find basic site information
Missing Schema.org markup — agents can't read structured data
JavaScript-only rendering — agents see a blank page
No AI-friendly protocols — you don't exist in the agentic web

The average website scores below 30/100 on AI agent readiness.

AgentCheck runs 18+ automated checks across 5 categories, shows which LLMs can find you, and gives you copy-paste code to fix everything.

Which AI Agents We Check

AI Agent	What We Test
ChatGPT (OpenAI)	llms.txt, ai.txt, structured data, protocol readiness
Claude (Anthropic)	Protocol readiness, agent accessibility, content structure
Perplexity	Structured data quality, trust signals, indexability
Gemini (Google)	Schema.org markup, crawl directives, structured data
Copilot (Microsoft)	AI protocols, structured data completeness
Google AI Overview	Schema markup, trust signals, crawl readiness

Each scan produces a per-agent visibility grid: see exactly which AI can find you and which can't.

What It Checks

Category	Points	What It Measures
Protocol Readiness	20	UCP endpoint, `ai.txt`, `llms.txt`, `robots.txt` AI bot rules
Structured Data Quality	25	JSON-LD, site-type Schema (Product, Article, Restaurant, etc.), Organization, Breadcrumbs
Agent Accessibility	20	TTFB, JavaScript dependency, API/feed availability, URL structure
Conversion Readiness	20	CTAs, contact forms, booking/checkout, site-type conversion signals
Trust Signals	15	Aggregate ratings, HTTPS + security headers, contact info

Every check produces a pass, warn, or fail with a specific fix suggestion and code snippet.

How It Works

1. Enter any website URL        →  agentcheck.site
2. We auto-detect site type     →  e-commerce, blog, SaaS, restaurant, etc.
3. We run 18+ automated checks  →  ~30 seconds
4. See which AI agents find you →  ChatGPT ✅, Claude ❌, Perplexity ✅ ...
5. Get your score + fix report  →  0-100 score, prioritized fixes with code

Example Output

Domain:     example-store.com
Type:       E-Commerce (auto-detected)
Score:      28 / 100  —  Grade: E
Visibility: Mostly Hidden

  ChatGPT     ❌  Missing llms.txt
  Claude      ❌  Protocol readiness too low
  Perplexity  ⚠️  Weak structured data
  Gemini      ❌  Schema markup needed
  Copilot     ❌  Missing AI protocols
  AI Overview ❌  Trust signals too low

Top Fixes:
  ✗ CRITICAL  No Product Schema found            (+6 pts)
  ✗ CRITICAL  No JSON-LD structured data         (+5 pts)
  ⚠ WARNING   No ai.txt file                    (+4 pts)
  ⚠ WARNING   Content requires JavaScript        (+5 pts)

Features

Free

AI readiness score (0-100) with letter grade
Per-agent visibility grid (ChatGPT, Claude, Perplexity, Gemini, Copilot, AI Overview)
Detailed report with 18+ checks across 5 categories
Fix suggestions with code snippets
Shareable report & badge

Fix Files ($9 one-time)

Ready-to-deploy llms.txt, ai.txt, Schema.org patches
robots.txt AI agent rules
Step-by-step implementation guide

Pro ($29/month)

Hosted AI Files — we serve your llms.txt & ai.txt for you
Agent Simulator — watch how ChatGPT navigates your site step-by-step
AI Mention Tracking — see how often LLMs cite your site
Crawler Ping — notify AI crawlers when you update (3/day)
Competitor Comparison — benchmark AI visibility vs rivals
Weekly Monitoring — alerts when your score changes
Score History — track improvements over time
Unlimited fix file downloads

Quick Start

Try It Online

agentcheck.site — free, no signup required.

Run Locally

# Clone
git clone https://github.com/lennystepn-hue/agentcheck.git
cd agentready

# Backend
cd backend
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

# Frontend (new terminal)
cd frontend
npm install
npm run dev

Open http://localhost:5173 and scan your first domain.

Docker (Production)

# Copy and configure environment
cp backend/.env.example backend/.env
# Edit .env with your keys

# Build and run
docker compose up --build -d

Tech Stack

Layer	Technology
Backend	Python 3.12, FastAPI, httpx, BeautifulSoup, aiosqlite
Frontend	Vue 3, Vite, Tailwind CSS, Composition API
Database	SQLite (persistent volume in Docker)
Auth	JWT (bcrypt + python-jose)
Payments	Stripe Checkout + Billing Portal
Deployment	Docker Compose, Nginx, Let's Encrypt SSL
Monitoring	APScheduler for weekly re-scans

API Reference

All endpoints are available at https://agentcheck.site/api/.

Public Endpoints

Method	Endpoint	Description
`GET`	`/api/health`	Health check
`POST`	`/api/scan`	Start a new scan
`GET`	`/api/scan/{id}`	Get scan status/result

Authenticated Endpoints

Method	Endpoint	Description
`POST`	`/api/auth/register`	Create account
`POST`	`/api/auth/login`	Sign in
`GET`	`/api/auth/me`	Get current user
`GET`	`/api/user/scans`	List user's scans
`GET`	`/api/scan/{id}/fixes/download`	Download fix ZIP (paid)
`POST`	`/api/compare`	Compare up to 4 domains (Pro)
`GET`	`/api/history/{domain}`	Score history (Pro)
`GET`	`/api/monitoring`	List monitored domains (Pro)
`POST`	`/api/monitoring`	Add domain to monitoring (Pro)
`POST`	`/api/hosted-files/activate`	Activate hosted AI files (Pro)
`POST`	`/api/crawler-ping`	Ping AI crawlers (Pro, 3/day)
`POST`	`/api/simulate/{scanId}`	Run agent simulator (Pro)
`POST`	`/api/checkout/session`	Create Stripe checkout
`POST`	`/api/billing/portal`	Stripe billing portal

Scan Example

curl -X POST https://agentcheck.site/api/scan \
  -H "Content-Type: application/json" \
  -d '{"domain": "your-website.com"}'

{
  "scan_id": "a1b2c3d4-...",
  "status": "pending",
  "message": "Scan started for your-website.com."
}

Fix Files — What You Get

File	Description
`llms.txt`	LLM context file describing your site — tailored to your site type
`ai.txt`	AI agent instructions with capabilities and endpoints
`robots_txt_additions.txt`	Rules for GPTBot, ClaudeBot, PerplexityBot, Google-Extended
`schema_*.jsonld`	Schema.org template matching your site type (Product, Article, Restaurant, etc.)
`schema_organization.jsonld`	Organization Schema for your business
`implementation_guide.md`	Step-by-step deploy guide (Shopify, WooCommerce, WordPress, custom)

Self-Hosting

AgentCheck is fully open source and designed to be self-hosted.

Requirements

VPS with 2+ GB RAM (Hetzner CX22 recommended, ~€4.50/month)
Docker + Docker Compose
Domain with DNS pointing to your server

Environment Variables

JWT_SECRET=<generate: python -c "import secrets; print(secrets.token_urlsafe(32))">
STRIPE_SECRET_KEY=sk_live_...
STRIPE_WEBHOOK_SECRET=whsec_...
STRIPE_PRICE_FIX_FILES=price_...
STRIPE_PRICE_PRO=price_...
FRONTEND_URL=https://your-domain.com
DATABASE_PATH=/data/agentcheck.db

Deploy

ssh root@your-server
git clone https://github.com/lennystepn-hue/agentcheck.git /opt/agentready
cd /opt/agentready
cp backend/.env.example backend/.env
# Edit backend/.env with your keys
bash deploy.sh

Project Structure

agentready/
├── backend/
│   ├── main.py                 # FastAPI app + all endpoints
│   ├── auth.py                 # JWT authentication
│   ├── payments.py             # Stripe integration
│   ├── monitoring.py           # Weekly re-scan scheduler
│   ├── fix_generator.py        # Tailored fix file generation
│   ├── db.py                   # SQLite database layer
│   ├── scanner/
│   │   ├── orchestrator.py     # Scan coordination
│   │   ├── protocol_checks.py  # UCP, ai.txt, llms.txt, robots.txt
│   │   ├── schema_checks.py    # JSON-LD, site-type Schema checks
│   │   ├── site_detector.py    # Auto-detect website type
│   │   ├── accessibility.py    # TTFB, JS dependency, API, URLs
│   │   ├── transaction.py      # Conversion readiness (site-type-aware)
│   │   ├── trust.py            # Ratings, security, contact
│   │   ├── discovery.py        # Product page finder
│   │   └── models.py           # Data models
│   └── snippets/               # Fix code templates
├── frontend/
│   ├── src/
│   │   ├── pages/              # Landing, Report, Dashboard, Pricing, ...
│   │   ├── components/         # ScoreCircle, CategoryBar, FixCard, ...
│   │   ├── auth.js             # Auth state management
│   │   ├── api.js              # API client
│   │   └── router.js           # Routes + guards
│   └── tailwind.config.js
├── docker-compose.yml          # Production deployment
├── nginx.conf                  # Reverse proxy + SSL
└── deploy.sh                   # One-command server setup

Related Standards & Protocols

AgentCheck evaluates compatibility with these emerging AI standards:

llms.txt — Markdown files that help LLMs understand your site
ai.txt — AI agent instruction files (like robots.txt for AI)
Schema.org — Structured data vocabulary for products, articles, businesses
UCP — Universal Commerce Protocol for programmatic AI commerce
robots.txt AI directives — Crawl rules for GPTBot, ClaudeBot, PerplexityBot, Google-Extended

FAQ

Is this free? Yes. The scanner and reports are free forever. Fix files and monitoring are paid.

What is GEO (Generative Engine Optimization)? GEO is the practice of optimizing your website to be discovered by AI agents and generative search engines. AgentCheck is a free GEO scanner.

How accurate is the score? The score measures technical readiness signals that AI agents look for. Sites with higher scores are significantly more likely to be found and recommended by ChatGPT, Claude, Perplexity, and other AI agents.

Which AI agents does this cover? ChatGPT (OpenAI), Claude (Anthropic), Perplexity, Gemini (Google), Copilot (Microsoft), Google AI Overview, and any agent that follows standard web protocols.

How is this different from SEO tools? Traditional SEO optimizes for Google crawlers. AgentCheck focuses on GEO and LLM SEO — optimizing for AI agents which read structured data, llms.txt, ai.txt, and machine-readable formats that search crawlers ignore.

Can I self-host this? Yes. AgentCheck is MIT licensed and fully self-hostable. See the Self-Hosting section.

What types of websites does this support? Auto-detects your site type: e-commerce, SaaS, blogs, restaurants, local businesses, professional services, portfolios, and more.

Contributing

Contributions are welcome. Please open an issue first to discuss what you'd like to change.

# Fork + clone
git clone https://github.com/YOUR_USERNAME/agentready.git
cd agentready

# Install
cd backend && pip install -r requirements.txt
cd ../frontend && npm install

# Run
cd ../backend && uvicorn main:app --reload &
cd ../frontend && npm run dev

Keywords

AI agent readiness scanner GEO tool generative engine optimization LLM SEO AI visibility checker ChatGPT SEO AI search optimization llms.txt generator ai.txt checker Schema.org validator AI readiness score website AI scanner Perplexity SEO Claude AI discovery Copilot optimization Google AI Overview AEO AI engine optimization get found by AI AI commerce agentic web agent-ready website

License

MIT — Lenny Enderle, 2026

Scan your website now — free, 30 seconds, no signup.

Built by Lenny Enderle

Name		Name	Last commit message	Last commit date
Latest commit History 62 Commits
.claude/skills		.claude/skills
backend		backend
docs/superpowers		docs/superpowers
frontend		frontend
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
card.html		card.html
deploy.sh		deploy.sh
docker-compose.yml		docker-compose.yml
make_twitter_card.py		make_twitter_card.py
nginx.conf		nginx.conf
screenshot_card.py		screenshot_card.py

Folders and files

Latest commit

History

Repository files navigation

AgentCheck

AI is recommending your competitors. Not you.

The Problem

Which AI Agents We Check

What It Checks

How It Works

Example Output

Features

Free

Fix Files ($9 one-time)

Pro ($29/month)

Quick Start

Try It Online

Run Locally

Docker (Production)

Tech Stack

API Reference

Public Endpoints

Authenticated Endpoints

Scan Example

Fix Files — What You Get

Self-Hosting

Requirements

Environment Variables

Deploy

Project Structure

Related Standards & Protocols

FAQ

Contributing

Keywords

License

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages