CrawlCheck
Data · census · weekly

Agentic web census, week 2026-W45

Sealed 2026-11-09. The same frozen sets of the most-visited sites, measured on a schedule from outside: what they tell and serve AI agents this week against last week. No site is named.

top 1,000, daily

Run 38 (finished 2026-11-08), 991 sites, compared with run 31 (2026-11-01).

Biggest moves: Of A2A cards naming a url, the url answers (any non-5xx, not an HTML page) -50 pt (0 gained, 1 lost); Homepage answers CrawlCheck's census fetch (2xx after redirects) +2.8 pt (38 gained, 10 lost); Homepage refuses CrawlCheck's census fetch (400, 401, 403 or 429): blocked or rate-limited, not down -1.8 pt (5 gained, 23 lost); No website answers at that name (5xx, 530, timeout or no answer) -1.2 pt (3 gained, 15 lost); GPTBot refused where the browser got 2xx -1.2 pt (3 gained, 9 lost).

FigureThis weekChange, same sitesSites +/−
Homepage answers CrawlCheck's census fetch (2xx after redirects)
of 991
54.2%+2.8 pt+38 / −10
Homepage refuses CrawlCheck's census fetch (400, 401, 403 or 429): blocked or rate-limited, not down
of 991
10.5%-1.8 pt+5 / −23
No website answers at that name (5xx, 530, timeout or no answer)
of 991
31.4%-1.2 pt+3 / −15
Serves a robots.txt (200, not HTML)
of 537
77.3%+1.2 pt+8 / −2
robots.txt names a sitemap
of 415
71.3%+0.3 pt+1 / −0
robots.txt carries a Content-Signal line
of 415
4.6%0 pt+0 / −0
Content-Signal says ai-train=no
of 415
3.1%0 pt+0 / −0
Content-Signal says ai-input=no or search=no
of 415
1%0 pt+0 / −0
robots.txt names the AI-training policy token (__usage_ai-training, IETF AIPREF draft)
of 414
1%——
robots.txt states a preference against AI training on / with that token
of 414
1%——
robots.txt closes / to at least one of the 12 AI agents
of 415
24.3%+0.3 pt+1 / −0
robots.txt closes / to GPTBot
of 415
15.4%0 pt+0 / −0
robots.txt closes / to OAI-SearchBot
of 415
7%+0.3 pt+1 / −0
robots.txt closes / to ChatGPT-User
of 415
8.9%0 pt+0 / −0
robots.txt closes / to ClaudeBot
of 415
14.7%0 pt+0 / −0
robots.txt closes / to Claude-SearchBot
of 415
8.2%+0.3 pt+1 / −0
robots.txt closes / to PerplexityBot
of 415
11.1%+0.3 pt+1 / −0
robots.txt closes / to Google-Extended
of 415
13%0 pt+0 / −0
robots.txt closes / to Applebot-Extended
of 415
11.8%0 pt+0 / −0
robots.txt closes / to CCBot
of 415
18.8%0 pt+0 / −0
robots.txt closes / to Bytespider
of 415
18.3%0 pt+0 / −0
robots.txt closes / to Meta-ExternalAgent
of 415
14.5%+0.3 pt+1 / −0
robots.txt closes / to Amazonbot
of 415
13.3%0 pt+0 / −0
Serves /llms.txt
of 537
22.9%+1.2 pt+7 / −1
Serves /agents.md
of 537
3%+0.6 pt+3 / −0
Publishes an agent-skills index
of 537
2.6%0 pt+0 / −0
Publishes an MCP server card
of 537
2.4%0 pt+0 / −0
Publishes an API catalog (RFC 9727)
of 537
2.4%+0.2 pt+2 / −1
Publishes an A2A agent card
of 537
0.6%0 pt+0 / −0
Declares any agent capability (any of the four above)
of 537
3.2%-0.2 pt+0 / −1
At least one AI identity is refused (non-2xx) where the browser got 2xx
of 537
20.9%-0.6 pt+7 / −10
At least one AI identity gets under 70% of the browser's bytes (both 2xx)
of 537
3.7%0 pt+3 / −3
Names an MCP endpoint (server card, llms.txt, agents.md, a homepage link, or /mcp answers like one)
of 537
10.8%+0.6 pt+3 / −0
An MCP endpoint completes the handshake (initialize and tools/list; no tool called)
of 537
1.7%0 pt+1 / −1
Of sites naming an MCP endpoint, one completes the handshake or asks for credentials
of 58
79.3%0 pt+0 / −0
The A2A agent card names an endpoint url
of 537
0.4%0 pt+0 / −0
Of A2A cards naming a url, the url answers (any non-5xx, not an HTML page)
of 2
0%-50 pt+0 / −1
Homepage carries WebMCP (navigator.modelContext, a WebMCP bridge, or an mcp+json block)
of 537
0.2%0 pt+0 / −0
Listed by Google Safe Browsing (all or some pages): browsers show a red warning
of 357
0.8%——
Listed as malware or phishing by the Cloudflare security resolver
of 550
0%——
On the browsers' HSTS preload list (HTTPS-only from the first visit)
of 550
16.5%——
Homepage loads scripts or styles from another domain
of 550
63.3%——
Homepage loads scripts or styles from 5 or more other domains
of 550
5.6%——
Pins at least one outside script or style with integrity=
of 348
9.8%——
Loads a library from a public CDN without integrity=
of 550
8.5%——
Still loads scripts from a CDN taken over in 2024 (polyfill.io and its sister CDNs)
of 550
0%——
Homepage shows hidden spam links or cryptominer code
of 550
0%——
GPTBot refused where the browser got 2xx
of 537
18.1%-1.2 pt+3 / −9
ClaudeBot refused where the browser got 2xx
of 537
17.7%-0.2 pt+5 / −6
OAI-SearchBot refused where the browser got 2xx
of 537
13.8%0 pt+5 / −5
PerplexityBot refused where the browser got 2xx
of 537
15.3%+0.6 pt+8 / −5

top 10,000, daily

Run 12 (finished 2026-11-08), 9776 sites, compared with run 5 (2026-10-31).

Biggest moves: PerplexityBot refused where the browser got 2xx +1.6 pt (113 gained, 27 lost); At least one AI identity gets under 70% of the browser's bytes (both 2xx) -0.8 pt (7 gained, 51 lost); Homepage answers CrawlCheck's census fetch (2xx after redirects) +0.6 pt (191 gained, 132 lost); No website answers at that name (5xx, 530, timeout or no answer) -0.6 pt (71 gained, 125 lost); Serves /llms.txt +0.5 pt (32 gained, 7 lost).

FigureThis weekChange, same sitesSites +/−
Homepage answers CrawlCheck's census fetch (2xx after redirects)
of 9776
58.1%+0.6 pt+191 / −132
Homepage refuses CrawlCheck's census fetch (400, 401, 403 or 429): blocked or rate-limited, not down
of 9776
13.5%-0.1 pt+77 / −91
No website answers at that name (5xx, 530, timeout or no answer)
of 9776
26.3%-0.6 pt+71 / −125
Serves a robots.txt (200, not HTML)
of 5680
81.9%+0.1 pt+36 / −31
robots.txt names a sitemap
of 4651
70.2%+0.3 pt+13 / −0
robots.txt carries a Content-Signal line
of 4651
6.3%+0.2 pt+10 / −2
Content-Signal says ai-train=no
of 4651
4.9%+0.1 pt+6 / −3
Content-Signal says ai-input=no or search=no
of 4651
0.5%0 pt+0 / −1
robots.txt names the AI-training policy token (__usage_ai-training, IETF AIPREF draft)
of 4647
0.3%——
robots.txt states a preference against AI training on / with that token
of 4647
0.3%——
robots.txt closes / to at least one of the 12 AI agents
of 4651
22%+0.1 pt+10 / −5
robots.txt closes / to GPTBot
of 4651
16.3%+0.1 pt+9 / −3
robots.txt closes / to OAI-SearchBot
of 4651
6.1%+0.1 pt+5 / −2
robots.txt closes / to ChatGPT-User
of 4651
8%0 pt+2 / −3
robots.txt closes / to ClaudeBot
of 4651
15%+0.1 pt+8 / −5
robots.txt closes / to Claude-SearchBot
of 4651
5.7%+0.1 pt+5 / −2
robots.txt closes / to PerplexityBot
of 4651
8.9%-0.2 pt+3 / −14
robots.txt closes / to Google-Extended
of 4651
13.4%+0.1 pt+9 / −5
robots.txt closes / to Applebot-Extended
of 4651
13.1%+0.2 pt+12 / −4
robots.txt closes / to CCBot
of 4651
17.1%0 pt+7 / −8
robots.txt closes / to Bytespider
of 4651
17%+0.1 pt+8 / −4
robots.txt closes / to Meta-ExternalAgent
of 4651
13.9%+0.2 pt+13 / −4
robots.txt closes / to Amazonbot
of 4651
13.4%+0.2 pt+11 / −3
Serves /llms.txt
of 5680
17%+0.5 pt+32 / −7
Serves /agents.md
of 5680
1.5%+0.1 pt+7 / −0
Publishes an agent-skills index
of 5680
1%0 pt+2 / −0
Publishes an MCP server card
of 5680
1.1%+0.1 pt+5 / −0
Publishes an API catalog (RFC 9727)
of 5680
1.2%+0.1 pt+5 / −0
Publishes an A2A agent card
of 5680
0.3%0 pt+2 / −0
Declares any agent capability (any of the four above)
of 5680
1.4%+0.1 pt+4 / −0
At least one AI identity is refused (non-2xx) where the browser got 2xx
of 5680
23%-0.4 pt+59 / −79
At least one AI identity gets under 70% of the browser's bytes (both 2xx)
of 5680
2.3%-0.8 pt+7 / −51
Names an MCP endpoint (server card, llms.txt, agents.md, a homepage link, or /mcp answers like one)
of 5680
5.8%+0.1 pt+10 / −2
An MCP endpoint completes the handshake (initialize and tools/list; no tool called)
of 5680
0.9%0 pt+3 / −1
Of sites naming an MCP endpoint, one completes the handshake or asks for credentials
of 327
78.9%0 pt+1 / −1
The A2A agent card names an endpoint url
of 5680
0.1%0 pt+1 / −0
Of A2A cards naming a url, the url answers (any non-5xx, not an HTML page)
of 8
37.5%0 pt+0 / −0
Homepage carries WebMCP (navigator.modelContext, a WebMCP bridge, or an mcp+json block)
of 5680
0.1%0 pt+0 / −0
Listed by Google Safe Browsing (all or some pages): browsers show a red warning
of 4614
0.9%——
Listed as malware or phishing by the Cloudflare security resolver
of 5698
0.3%——
On the browsers' HSTS preload list (HTTPS-only from the first visit)
of 5698
8%——
Homepage loads scripts or styles from another domain
of 5698
67.5%——
Homepage loads scripts or styles from 5 or more other domains
of 5698
10.9%——
Pins at least one outside script or style with integrity=
of 3845
10.1%——
Loads a library from a public CDN without integrity=
of 5698
13.3%——
Still loads scripts from a CDN taken over in 2024 (polyfill.io and its sister CDNs)
of 5698
0.1%——
Homepage shows hidden spam links or cryptominer code
of 5698
0%——
GPTBot refused where the browser got 2xx
of 5680
19.2%0 pt+37 / −38
ClaudeBot refused where the browser got 2xx
of 5680
20%0 pt+52 / −54
OAI-SearchBot refused where the browser got 2xx
of 5680
12.4%-0.4 pt+25 / −46
PerplexityBot refused where the browser got 2xx
of 5680
14.5%+1.6 pt+113 / −27

Each frame compares the last run that finished in this ISO week (UTC) with the last run that finished before it. Change is like-for-like: points = (sites gained - sites lost) / sites compared, counting only sites measured in both runs and inside the figure's denominator both times, so a site that did not answer in one run cannot move a figure. points_all is the plain difference of the two percentages, and denominator_entered / denominator_left count the sites that moved in or out of a figure's base (often a site that did not answer one run). No site is named; per-site records: https://crawlcheck.io/api/census/records.

Signed JSON: /api/v1/census/weekly?week=2026-W45 · feed: Atom · method: /api/census. Editions: 2026-W46 · 2026-W45 · 2026-W44 · 2026-W43 · 2026-W42 · 2026-W41 · 2026-W40.

Check it yourself

Every number here comes from the signed JSON at /api/v1/census/weekly; each copy carries its own sha256 and an Ed25519 signature by key 1hHmE3fo1rg5…. Verify it with no call to us:

node crawlcheck-verify.mjs --online census-weekly.json

The verifier: /crawlcheck-verify.mjs, zero dependencies.

Editions

One sealed edition per ISO week, never edited. Feed: Atom · all instruments: /api/v1/instruments.