Skip to main content

Find the gaps that keep your site out of AI answers.

Buyers ask ChatGPT, Perplexity, Claude, and Gemini before they click a search result. GeoReady checks whether your URL can be reached, understood, and cited by those answer engines, then gives you a 0–100 score and prioritized fixes across 8 signal categories.

Free report, no account required. POST /api/audit

809
GitHub stars
74.0k
downloads/mo
2.0k
audits run

live · GET /api/stats

The eight gates every audit runs

models/config.py · SCORING · Σ = 100 pts

  1. 01 Robots.txt 18 pts robots queued

    Whether citation crawlers are allowed to fetch the site at all.

  2. 02 llms.txt 18 pts llms queued

    Presence and depth of the root-level orientation file, including llms-full.txt.

  3. 03 Schema JSON-LD 16 pts schema queued

    Valid structured data: Organization, WebSite, Article, FAQ.

  4. 04 Meta tags 14 pts meta queued

    Title, description, canonical, and Open Graph completeness.

  5. 05 Content 12 pts content queued

    Heading hierarchy, word count, lists, front-loaded answers.

  6. 06 Signals 6 pts signals queued

    Declared language, feed availability, freshness date.

  7. 07 AI discovery 6 pts ai_discovery queued

    The .well-known/ai.txt and /ai/*.json endpoints.

  8. 08 Brand & entity 10 pts brand_entity queued

    Naming coherence, Knowledge Graph readiness, topic authority.

AI SEO is not a checklist. It is a retrieval path.

Traditional SEO asks whether a page can rank. AI SEO asks whether an answer engine can reach it, parse it, trust it, and cite it.

  1. 01Access

    Crawler permissions, HTTP behavior, and server-rendered content decide whether AI systems can reach the page.

  2. 02Meaning

    llms.txt, schema, headings, and entity consistency help the model understand what the page is about.

  3. 03Citation

    Self-contained answers and factual density make the page easier to quote inside generated responses.

How GeoReady works

1Enter a URL

Submit any public website address. The tool resolves and validates the URL, then fetches the page source for analysis.

2Get a GEO score

The scan evaluates eight signal categories and returns a 0–100 score with a per-category breakdown. These are the published weights — the same numbers the open-source engine uses, so any score can be recomputed by hand:

Gate Category Points What it measures
01 Robots.txt robots 18 Whether citation crawlers are allowed to fetch the site at all.
02 llms.txt llms 18 Presence and depth of the root-level orientation file, including llms-full.txt.
03 Schema JSON-LD schema 16 Valid structured data: Organization, WebSite, Article, FAQ.
04 Meta tags meta 14 Title, description, canonical, and Open Graph completeness.
05 Content content 12 Heading hierarchy, word count, lists, front-loaded answers.
06 Signals signals 6 Declared language, feed availability, freshness date.
07 AI discovery ai_discovery 6 The .well-known/ai.txt and /ai/*.json endpoints.
08 Brand & entity brand_entity 10 Naming coherence, Knowledge Graph readiness, topic authority.

Bands: 86–100 excellent · 68–85 good · 36–67 foundation · 0–35 critical.

3Apply prioritized fixes

Each finding is ranked by impact and assigned a severity level. The report includes actionable recommendations with specific file changes and schema additions.

Run summary · what a typical site scores

exit 0

We measured the sites scored through GeoReady, and our data shows the average scores 54.7 out of 100, with a median of 57. Only 58% publish an llms.txt file. A score in the fifties is therefore ordinary rather than alarming: most sites leave a large share of their AI-visibility points unclaimed, and the gaps sit in signals a tool can find mechanically.

GeoReady benchmark, 1100+ audited sites, as of 17 September 2026. Latest audit per domain, aggregates only — the sample is self-selected, so read it as a baseline among sites that chose to measure. See the live benchmark.

Why trust this tool

  • Open source (MIT)

    github.com/Auriti-Labs/geo-optimizer-skill

    Full methodology, scoring weights, and source code on GitHub. No proprietary scoring.

    View on GitHub
  • Research-backed

    arXiv 2311.09735 · 2510.11438

    Scoring grounded in Princeton KDD 2024 and AutoGEO ICLR 2026 research on AI search retrieval.

    Read the research
  • Auditable scoring

    models/config.py · SCORING

    Every signal is traceable to a published rule. Score breakdowns show per-category contributions.

  • Machine-readable output

    geo audit --format json

    JSON format designed for CI/CD pipelines, integrations, and the GeoReady platform API.

One audit gives the baseline. Monitoring tells you when the baseline moves.

The free audit is a point-in-time diagnosis. GeoReady Pro turns the same scoring engine into an ongoing signal for teams and client work.

Free audit

run: once

GEO score, top recommendations, and a fast read on whether a URL is crawlable and understandable.

GeoReady Pro

schedule: weekly

Weekly monitoring, report history, regression alerts, PDF export, and API access.

Start with Pro

Frequently asked questions

What is Generative Engine Optimization?

GEO is the practice of making websites discoverable, readable, and citable by AI search engines such as ChatGPT, Perplexity, Claude, and Gemini. It focuses on structured data, content hierarchy, and technical signals that AI crawlers prioritize.

What does GeoReady check?

GeoReady audits 8 categories: robots.txt, llms.txt, JSON-LD schema, meta tags, content quality, technical signals, AI discovery paths, and brand entity signals. It also detects prompt injection patterns and negative signals through the same open-source GEO Optimizer engine. The CLI additionally offers Agent Access Audit (bot vs browser access simulation via `geo access`) and AI Crawler Activity Analytics (server log parsing via `geo logs`).

How is GEO different from traditional SEO?

Traditional SEO optimizes for keyword rankings and backlinks in conventional search engines. GEO optimizes for AI citation signals: structured data, clear content architecture, AI crawler permissions, and self-contained answer blocks. The two complement rather than replace each other.

Is the Pro plan available now?

Yes. Pick a plan and your account activates immediately. No sales call, no waitlist.

Is GeoReady open source?

The core GEO Optimizer engine is MIT licensed. Source code, scoring weights, and methodology are publicly available on GitHub. The CLI and the free web audit remain free; GeoReady Pro adds monitoring, history, alerts, exports, and team workflows.

Does the audit store submitted URLs?

URLs are processed in real time. No persistent storage of submitted URLs is declared unless explicitly configured in the backend. See the Privacy Policy for details.

AI search is already deciding which sites get cited.

Verdict · your site: not yet run

The signals are documented, the research is published, and the tool is free. GeoReady exists to make AI visibility auditable and accessible to every developer.