AI Search Readiness: measure structure, ship dry-run fix packs, rescore, seal with HMAC receipts. SEO tests if humans can find you on Google. AIO.GEO tests if AI agents can do business with you.
Every "AI" and "ChatGPT ranking" tool I tested was selling a number nobody can audit. Answer engines do not publish a stable SERP. That is the problem.
AIO.GEO measures what you can prove: structure, schema, JSON-LD, crawl access. Then it ships a dry-run patch and an HMAC receipt.
Try it in 20 seconds (no install):
npx --yes @aio-geo/cli@0.3.4 doctor
npx --yes @aio-geo/cli@0.3.4 audit yourdomain.com
You get a 5-pillar AI Search Readiness score, not an LLM rank.
Open-core (MIT): @aio-geo/cli, @aio-geo/sdk, @aio-geo/mcp-server, @aio-geo/action
Site: https://www.aiogeoprotocol.com
CLI: https://www.aiogeoprotocol.com/cli
npm: https://www.npmjs.com/package/@a...
We do not sell LLM rankings. Nobody can measure them honestly. We sell structure, accessibility, and receipts to prove it.
Report
"Receipts" is the part that jumps out most GEO/AEO advice right now is still guesswork dressed up as best practice, because there's no clean feedback loop like there is with traditional SEO. You can rank #1 in Google and still have no idea whether an AI answer engine ever surfaces your page, or why.
Ran into this firsthand recently pulled the new GSC Generative AI performance report expecting a real signal, and the data was messy enough (heavily skewed toward my own testing traffic, hard to separate from real exposure) that I couldn't actually trust it yet. So genuinely curious how you're defining a "fix" here is it structural stuff (schema, headings, answer-formatted content), or are you also weighing in on the actual writing/content clarity side, since a lot of AI-answer visibility seems to come down to how directly a page answers the implied question, not just how it's marked up?
@farrukh_ahmed8877 Spot on, Farrukh! The GSC Generative AI noise is exactly why everyone is throwing darts in the dark right now.
To answer your question: the "fix" covers both, but we separate them cleanly into two axes:
Cognitive (Readability & Content Clarity): Auditing how directly your content answers implied questions, structural schema, and RAG-readability so LLMs can parse and cite you without hallucinating.
Kinetic (Actionability & Code): Going beyond text to generate actual execution assets (llms.txt, api-catalog, MCP cards, auth.md) so AI agents can actually do business on your site (call APIs, auth, perform tasks), not just read it.
You get an exact diagnosis of what's blocking AI, the dry-run files to fix it on the spot, and a rescore to verify the improvement. Receipts over guesswork!
Report
@adrian_swish The cognitive/kinetic split is a useful distinction most people treat "AI visibility" as one blob. On the cognitive side is the fix usually structural (reordering, Q&A framing), or does the actual prose need to change at the sentence level too?
@farrukh_ahmed8877Structure ensures the AI search engine finds and retrieves your page's chunk. Prose clarity ensures the LLM understands and cites your facts accurately without hallucinating or making up numbers.
Report
Congrats on the launch! Focusing on verifiable technical foundations instead of arbitrary 'LLM ranks' is a refreshing approach to GEO.
How are you handling dynamically generated content or JavaScript-heavy frameworks when auditing the readiness score?
The dry-run and re-score workflow is a nice touch. Being able to see what a change would actually affect before applying it makes the whole audit process much easier to work with. Congrats on the launch!
AIO.GEO Protocol
"Receipts" is the part that jumps out most GEO/AEO advice right now is still guesswork dressed up as best practice, because there's no clean feedback loop like there is with traditional SEO. You can rank #1 in Google and still have no idea whether an AI answer engine ever surfaces your page, or why.
Ran into this firsthand recently pulled the new GSC Generative AI performance report expecting a real signal, and the data was messy enough (heavily skewed toward my own testing traffic, hard to separate from real exposure) that I couldn't actually trust it yet. So genuinely curious how you're defining a "fix" here is it structural stuff (schema, headings, answer-formatted content), or are you also weighing in on the actual writing/content clarity side, since a lot of AI-answer visibility seems to come down to how directly a page answers the implied question, not just how it's marked up?
AIO.GEO Protocol
@farrukh_ahmed8877 Spot on, Farrukh! The GSC Generative AI noise is exactly why everyone is throwing darts in the dark right now.
To answer your question: the "fix" covers both, but we separate them cleanly into two axes:
Cognitive (Readability & Content Clarity): Auditing how directly your content answers implied questions, structural schema, and RAG-readability so LLMs can parse and cite you without hallucinating.
Kinetic (Actionability & Code): Going beyond text to generate actual execution assets (llms.txt, api-catalog, MCP cards, auth.md) so AI agents can actually do business on your site (call APIs, auth, perform tasks), not just read it.
You get an exact diagnosis of what's blocking AI, the dry-run files to fix it on the spot, and a rescore to verify the improvement. Receipts over guesswork!
@adrian_swish The cognitive/kinetic split is a useful distinction most people treat "AI visibility" as one blob. On the cognitive side is the fix usually structural (reordering, Q&A framing), or does the actual prose need to change at the sentence level too?
AIO.GEO Protocol
@farrukh_ahmed8877 Structure ensures the AI search engine finds and retrieves your page's chunk. Prose clarity ensures the LLM understands and cites your facts accurately without hallucinating or making up numbers.
AIO.GEO Protocol
@hannesh thank you
Voquill
Congrats on the launch! I like that you can make the fixes and then rescore to see if they actually worked.
AIO.GEO Protocol
@henry_habib thank you Henry!
AIO.GEO Protocol
@henry_habib Thank you Henry
AIO.GEO Protocol
@etiennegarcia Thank you Etienne !