Public audit stats

Long-tail view of the corpus. Numbers refresh every 5 minutes.

Total sites tracked
13744
New (30d)
21
Last refresh
2026-09-18 23:46 UTC

Score distribution across hosts

One score per host: the latest valid run under the current rubric, over the whole corpus. The table below counts something different — individual runs of the last 30 days, under every rubric version — so the two totals are not meant to agree.

025507510011,071 scored hosts · rubric v7 · submitted URLs, not a curated list1,259 · 11.3%median 27
These are URLs people submitted themselves, not a curated list of API documentation. The distribution describes what was submitted, not the market: much of it is not API documentation at all and scores near zero. SVG

Score distribution (last 30 days)

BandCountShareAvg
Good 133 2% 73.8
Fair 818 13% 58.1
Weak 2461 38% 38.7
Poor 3039 47% 17.2

Distribution buckets from the materialised view: Good ≥ 70 · Fair 50–69 · Weak 30–49 · Poor < 30. (Score-tag colours elsewhere on this page use the brand-book 2-band scheme: pass ≥ 90 · fail < 50 · neutral 50–89.)

Where the points are earned

SCORING 0EARNED SHARE OF EACH CATEGORY'S BUDGET0%25%50%75%100%ADiscovery · 14 pts31.8%29.1%BPage artifacts · 21 pts36.5%10.6%CAPI contract · 17 pts16.3%7.3%DContent · 23 pts21.0%21.3%ERendering & hygiene · 21 pts42.1%6.7%FAgent capability · 4 pts0.7%98.8%11,120 entry URLs · rubric v7 · submitted URLs, not a curated list
Each bar is the share of that category's own budget, and the budgets differ, so a short bar in a small category is not a large loss. The right-hand column counts sites that earned nothing there, which includes criteria that did not apply to them. SVG

Measurement 1270 of 7721

Runs of the last 30 days by whether the audit obtained enough responses to score the site. Every run that did not is excluded from every table on this page.

ClassCountShare
Scored645183.6%
Refused or sign-in4956.4%
No response4505.8%
Partial read3254.2%

A non-scored run says the audit could not read enough of the documentation — it is not a judgement about the site. Some of it is self-inflicted: our own re-audit sweep is refused about three times as often as a one-off visitor request.

Top organisations (eTLD+1)

OrgSitesAvg bestMax
voidly.ai 2 80.5 82
z.ai 2 75.5 76
docker.com 2 70.0 73
gemini.com 2 66.0 66
nexudus.com 2 64.0 66
datadoghq.com 2 61.5 62
cloudflare.com 3 61.0 71
emergingtravel.com 4 60.5 80
wtfip.io 2 60.0 60
amplitude.com 2 59.5 67

Orgs with ≥2 audited subdomains, ranked by avg of best scores.

TLD distribution

TLDSites
.com8189
.io1008
.org534
.ai384
.net331
.ru221
.co212
.de142
.dev124
.app98
.gov93
.com.br88
.eu62
.cn61
.me51
.nl49
.cloud46
.jp46
.fr43
.in43
.se42
.vn42
.ca40
.pl39
.tech39
.com.au38
.xyz38
.co.uk37
.fi36
.it34
.cz32
.ch31
.no29
.network26
.es25
.finance24
.id24
.edu23
.dk22
.sh22
.co.kr21
.cl19
.so19
.tv19
.co.za18
.at17
.be17
.info17
.gr16
.ee15

JSON: /stats · /summary · ← browse