Agent readiness, measured from outside
Type your domain and 15 checks run against it in half a minute, telling you which step an unattended agent dies on: finding you, reading your docs, creating an account, getting a key, or the first API call. When a developer says “add file uploads”, an agent picks the provider and writes the integration. It never files a support ticket to tell you it gave up.
Of 141 vendors we can compare today (we hold 177; the rest are waiting for a rescan), measured under formula 9.30 while the scanner runs 9.31, so a scan you run today can disagree with the row below it, 75 serve a signup form that renders nothing without JavaScript, which an agent reads as a page with no way in. Outright refusals aimed at agents are rarer than the noise around them suggests, and we say so rather than counting every 403 our data centre collects.
Your scan gets a permanent link you can forward. It is never added to the published corpus, and we do not post it anywhere.
Free. No account. Reads only what you publish.
Eighteen runs in isolated copies of a real application, across the four categories we have run agents in. In the three that need an account, every run stopped at the signup form and said so in its own words. In the fourth no account was needed and the barrier still appeared, earlier: a required licence key struck vendors off during dependency research, before their product was opened.
In one study the cheaper model fetched no external source in ten runs and said so: own knowledge only. In another, where the choice turned on a licence, every run fetched sources including all of the cheaper ones. A decision that cannot be answered from memory is what makes your documentation get read.
The app already had a session cookie. Adopting that provider meant running a second identity system just to upload a file, and three separate runs rejected it in almost the same words.
One provider was never selected, yet was considered and dismissed in nineteen of twenty runs, always because it assumes a framework the project did not use. One documentation chapter fixes it.
Thirty-eight runs across five studies, with the limits of each one stated. Read the method
138 domains · every result is a live link
Ordered by the share of measurable points each domain earned, which is why a smaller number can sit above a larger one. These are real scans, not examples. Each vendor can open its own scorecard, reproduce every check and tell us where we are wrong. That offer is the whole point of publishing the formula. Scores are out of the points we could measure on each domain, not out of the paper maximum: a site that refuses our requests scores against a smaller denominator, not a worse number.
A score marked * comes from a domain where too much of the card was unreadable to compare it with the rest: usually the site refused us, sometimes we could not reach the pages a check needs, and either way the fault can be ours. Checks that simply do not apply, like an account on a library you install rather than sign up for, are not counted against it. Those rows sit at the bottom of their category whatever the percentage says, so that blocking the scanner cannot be a way to lead it.
6 more in this category, all of them in the data.
3 more in this category, all of them in the data.
2 more in this category, all of them in the data.
2 more in this category, all of them in the data.
2 more in this category, all of them in the data.
2 more in this category, all of them in the data.
2 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
1 more in this category, all of them in the data.
Most scanners stop at whether you publish a file. The interesting failures happen further down the funnel, where an agent has already chosen you and still cannot finish.
A dashed column is a stage we could not measure, not a stage you failed.
What the five stages look like on one real domain
name.com 11 / 15 · current leader in domain registration and dns
What a machine can see
15 deterministic checks across the five stages, 17 points on paper. Published formula, reproducible result, no model involved. Runs in under a minute.
You are scored out of what we could measure on your domain, not out of 17. Across the 141 domains scanned here that averages 14 points, and only 1 could be measured in full.
Your scan gets a permanent link you can forward. It is never added to the published corpus, and we do not post it anywhere.
When it breaks, you hear it from us
An edge rule that starts refusing agents changes nothing a person sees in a browser, so the first sign is usually an integration that quietly stopped working. We rerun the checks weekly and write only when a verdict moves, naming what it says now and what it said before.
Once a month a real agent is asked the question your buyers ask, five times over, and you are told how many of the five named you and which provider was picked instead. It answers a question and signs up for nothing, so nothing of yours is touched. That half covers the 26 categories we measure, and if your product is not in one of them we tell you rather than quietly skip it.
Free while we are building it, and we will ask before it ever costs anything. No account, no card, and one link in every email stops it.
Want the agent runs once rather than every month? One report, ten runs, two tools, and you keep the transcripts.
Run by Krystian Gwizdała. The scanner, the formula, the agent runs and every number on this site are mine, and an audit is run by me rather than by a team you never meet. Four audits are published in full so you can see the work before deciding whether it is worth anything to you.