AI Audit Tool
Desktop tool for auditing website AI-readiness. Scans pages, evaluates 6 pillars (indexability, schema, DOM semantics, content, media, operational signals) and generates PDF reports with actionable recommendations.
01 Details
About the Project
AI Audit Tool (GEO Auditor) is a desktop application for SEO/GEO consultants and webmasters that analyzes how well websites are prepared for the era of AI search engines — ChatGPT, Perplexity, Google SGE, and other AI assistants.
Problem
AI assistants and search engines need structured, semantically clean data to reliably extract information about products, services, and companies. Most websites lack proper Schema.org markup, have unstructured content, and AI systems cannot reliably extract key facts from them.
Solution
AI Audit Tool performs a deep analysis of entire websites and evaluates their AI-readiness based on 6 pillars:
- Indexability & Access — robots.txt, meta directives, canonical URLs, HTTP status
- Structured Data Coverage — Schema.org types and their completeness
- Extractability & DOM Semantics — H1/H2 structure, semantic content, DOM quality
- Content Completeness — facts vs. fluff ratio, presence of critical fields
- Media & Alt Context — image ALT texts, media quality and size
- Operational Signals — sitemap, breadcrumbs, internal linking
Key Features
- Two modes — ⚡ Quick Snapshot (no AI, fast overview) and 🤖 AI Enhanced (with Gemini, deep analysis)
- Template clustering — two-phase pipeline that groups similar pages and analyzes only representatives (saves time and API calls)
- Agent Extraction Test — deterministic test of whether AI can extract key facts from the site
- Checkpoint system — audit can be stopped and resumed at any time
- PDF reports — professional PDFs with visualizations, scores, and actionable recommendations
- Implementation Checklist — Excel export with prioritized tasks (P0/P1/P2) for developers
- Email outreach — integrated module for cold outreach to potential clients
- ARES integration — company verification from Czech business registry
- Licensing system — Lemon Squeezy integration (Standard/Recommended/PRO plans)
Tech Stack
Python + Streamlit (UI), Playwright (JS rendering), BeautifulSoup + lxml (parsing), Google Gemini (AI analysis), FPDF2 (PDF generation), Plotly (visualization), SQLite (audit history).