About
An open standard, not a product
What this is, who made it, how it is run and how it relates to the companies behind it.
On this page
What this is
The AI Readiness Index is a published standard for how well a website can be found, read, trusted and used by machines: search crawlers, assistants and agents. It defines 72 tests, the 38 that are worth points and how those points add up to a score out of 100, and the evidence behind each. Anyone can read it, implement it, argue with it and propose changes to it.
It does not measure design quality, marketing or whether the content is any good. A beautiful site can score 12 and a plain one 94.
Who made it
The standard was created by TEN7, which maintains it together with outside contributors. TEN7 is a company. readinessindex.io is not a separate company: it is a product of TEN7, a scanner that implements the standard.
Neither TEN7 nor its product is the standard. This site is not an advertisement for either. They are linked so you can see where the standard came from, not to recommend them. A scanner by anyone else that follows the text is just as much an implementation, and a score from any of them can be checked against what is written here.
The line is drawn at the text. What a site has to do to score well, how the points are weighted and what counts as evidence are public and licensed for reuse. The software a particular company writes to measure a site against that text is that company's own and is not part of the standard.
How it is run
- Changes are made in public. Proposals, corrections and objections are issues and pull requests in the repository. Questions and open-ended ideas go to Discussions.
- Every change is recorded. The changelog says what changed in each version and why.
- A change to a score is a major version. Any change to a weight, a gate or the scoring model bumps the major version and says plainly that old and new numbers cannot be compared.
- Evidence decides what is scored. A test earns points only if something is documented using it, and every test is re-read every six months.
- Maintainers decide. Renumbering a test is done by the maintainers, together with anyone who implements it, in one pass, so that no implementation is left pointing at an ID that now means something else.
License
The text of the standard, the registry, the schema and the files on this site are published under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0). You may copy it, adapt it and build your own scanner from it, for any purpose including commercial ones, free and for good, as long as you credit TEN7 as the creator of the AI Readiness Index and license what you build from it the same way.
The license covers the standard, not its name. Anyone may say their tool implements the AI Readiness Index version 12.0; nobody may use the name to suggest that TEN7 or this project endorses a product.
Privacy
This site sets no cookies, loads no third-party scripts and fonts, and runs no analytics. It is static files served by GitHub Pages, which keeps its own access logs.