Model Visibility

People increasingly ask an assistant about an API instead of reading its documentation. So we ask three of them, on a schedule, about companies in this catalog — and check the answers against the artifacts those companies actually publish.

Last run 2026-09-28 · wave 1 · 65 providers · claude-opus-5 · gpt-5.5-2026-04-23 · gemini-flash-latest

What we measure

Every question is asked twice, and the two answers mean different things. With search enabled we learn what a model can find. With tools off we learn what it already knows. Those move on completely different clocks, so they are never pooled.

How it works

We ask closed questions, so an answer can be checked by a rule rather than by another model’s opinion:

• Does <company> publish a public OpenAPI specification?
• Does <company> operate an MCP server for its API?
• Does <company> offer a free tier or trial for its API?

Each answer is compared against that company’s own published artifacts, as recorded on its provider page. Where we do not hold the fact, nothing is scored — a gap in our profile is our backlog, not the model’s error. Saying “unknown” is counted separately and is not treated as a wrong answer: a model that declines to guess about a small company is behaving correctly, and a metric that punished that equally would be recommending confabulation.

How a company gets measured

Every provider in the exemplar band of the Kin Score is measured here, free. Not a sample of them — all of them. Reaching exemplar is what earns the measurement, and that is deliberate: we are one of very few parties who can tell a company what frontier assistants say about it and check whether that is true, because we hold verified ground truth about its artifacts. It costs a provider nothing but a score.

If you are below exemplar, the blueprint is the shortest route up — it lists every machine-readable document we publish, what each is worth on the rubric, and which of them cost nothing but a file. Fifty-eight of the 139 agent-readiness points come from publishing documents about an API you already have.

A company below exemplar can also ask to be measured. Those are marked requested and are counted separately, everywhere — a company that asks to be measured is not a random company, and averaging the two would move the published number for a reason that has nothing to do with models.

What this population is not. The exemplar band is the top ~1.1% of the catalog by construction, so nothing on this page is a statement about APIs in general. It describes these companies and no others. A stratified panel that could have supported a catalog-wide estimate was retired in September 2026; its measurements are kept, and its members rejoin here if they reach exemplar.

Worth saying plainly: we also sell the work that changes these numbers. A requested measurement is run on the same instrument as everything else, its result is never adjusted, and it is published whether it flatters the company or us. If that is not verifiable enough, the underlying data is linked at the bottom of this page.

The providers measured

Each row links to the provider page holding the artifacts every answer was checked against, so any claim here can be re-checked rather than taken on trust. A low score is a finding about the assistant, not about the company.

Reading this honestly

• A wave is a sample. One provider changing its mind is a few percentage points. Read “not observed in this sample”, never “the model does not know”.
• The three assistants are never averaged. They disagree about the same company, and the disagreement is the point — a company fixing one of them fixes it in a different place than the others.
• A model version change breaks the series. Two of the three are pinned to a dated build. One is a moving alias, so its line before and after a version change is two series, not one trend.
• We are one of the sources being measured. “Cited us” and “used our vocabulary” are numbers about our own reach, and we report them here rather than only in private because they are the least flattering numbers on the page.

The data behind this page is public. model-visibility.json carries every claim, what each assistant said, what we hold, the exact model ids and the capture timestamp — so nothing here has to be taken on trust.

If a result here is wrong about your company, that is worth telling us — the fastest fix is usually the artifact the model could not find. Claim or correct your listing →