AgenticWeb CensusCheck your site
EN

September 2026 · Lighthouse 13.4.0

Is your site ready for AI agents?

More and more AI assistants visit websites to do things for you. Enter your URL and see what they can do on it, what needs fixing, and how you compare with the rest of the web.

How the web is doing this month

Every figure over time

The average score of the 145,652 pages HTTP Archive measured with Lighthouse in September 2026. For your own site's score, use the form above.

Weak34.3%

Average score of the web · AWCS

Measured on a sample of 138,536 of the pages HTTP Archive audits.

What this score measures and how it is computed▲ +0.4
Average score of the web · AWCS over the last 5 months

30.2%34.3%

Read

What Read measures

Lighthouse builds the accessibility tree an agent would traverse and scores the page pass or fail. The denominator is only the pages the audit could evaluate: a not-applicable result or an error is excluded rather than counted as a failure, and the fraction that could be evaluated is published beside it as the coverage.

n=138,548 · coverage 95.1% · margin ±0.14 pp.

40% of the score
Read on its full scale

Weak31.8%

Read: 5 months, up▲ +1.1 pp
Stable

What Stable measures

Lighthouse's Cumulative Layout Shift audit, counted per page: a page passes when Lighthouse rates its layout shift as good. The denominator is the pages the audit scored. It matters to an agent that works the page the way a person does, by taking a screenshot and clicking where it saw a button: if the page has moved in between, it clicks something else.

n=138,551 · coverage 95.1% · margin ±0.14 pp.

25% of the score
Stable on its full scale

Good78.2%

Stable: 5 months, up▲ +0.4 pp
Guide

What Guide measures

The pages that pass the llms-txt audit: they serve a file and Lighthouse considers it valid. The denominator is every sampled page, so this is a fraction of the whole web and not of the sites that have a file. This is the figure the score uses, because a broken file does not guide an agent.

n=138,536 · coverage 95.1% · margin ±0.14 pp.

20% of the score
Guide on its full scale

Poor14.2%

Guide: 5 months, down▼ −0.4 pp
Act

What Act measures

A count of the sampled sites that register WebMCP tools that an agent can call. It is the audit that defines the Act dimension; the other WebMCP audits measure how good those tools are once they exist.

n=138,551 · coverage 95.1% · margin ±0.14 pp.

15% of the score
Act on its full scale

Very poor4.6%

Act: 5 months, down▼ −0.4 pp
Reachable

What Reachable measures

Read from the robots.txt of the sampled sites: of the AI agents we track, how many are not disallowed. It is gradual, from none to all of them, because a site that turns away one agent still lets the others in. A site with no robots.txt blocks nobody, so it counts as fully reachable, and that is what was measured. This is the only figure that multiplies the score instead of adding to it, because it is a precondition: if an agent cannot fetch the page, nothing else about the page matters. One caveat: a blanket Disallow under User-agent: * is not counted, because the corpus-wide measurement cannot see it either, so both are measured the same way.

n=145,652 · coverage 100% · margin ±0.14 pp.

multiplies the rest
Reachable on its full scale

Excellent95.8%

Reachable: 5 months, no significant change= +0.1 pp

Technologies

All 731 technologies

How to read these figures

The colour says what a bar represents, not how it scores: the accent is the one this page is about, grey is everything it is compared with, and the hollow bar is the web overall, which serves as the reference. Whether a figure is above or below the rest of the web is shown by the number and the arrow beside it.

Every figure here comes from a sample, not from counting every site, so it carries a margin: how much it could move if we had happened to sample different pages. A figure measured on a thousand pages has a wide margin; one measured on a hundred thousand has a narrow one. It matters for one thing: two figures that differ by less than the wider of their margins count as tied, even if one is slightly higher. That is why nothing here has a position number.

Every dimension adds up except one, which multiplies: how much of the site the AI agents we track can reach. It works that way because it is a precondition: if an agent cannot fetch the page, nothing else on it matters, so a site that turns away every agent scores zero however good the rest of it is. Blocking used to earn points here, on the reasoning that having decided was a sign of maturity; in a score that measures how ready a site is for AI agents, that was backwards. Readiness is a matter of fact, and whether you want to be ready is your call: a low score here records that you chose to be closed, and that choice is yours to make.

The census also measures analytics tags, payment processors and protocol features, whose figure describes their customers rather than themselves, which is why they are not on this list.

The average score of the sites that use each category of technology, best first.

CategoryAverage scoreTechnologies
Static site generator

Static site generator

Technologies
3
Average score
42.4%
Margin
±5.01
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Static site generator averages 42.4%42.4%
3
PaaS

PaaS

Technologies
19
Average score
35%
Margin
±5.12
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

PaaS averages 35%35%
19
Page builders

Page builders

Technologies
20
Average score
34.5%
Margin
±5.3
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Page builders averages 34.5%34.5%
20
JavaScript frameworks

JavaScript frameworks

Technologies
32
Average score
34%
Margin
±4.78
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

JavaScript frameworks averages 34%34%
32
CMS

CMS

Technologies
34
Average score
33.2%
Margin
±5.25
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

CMS averages 33.2%33.2%
34
CDN

CDN

Technologies
23
Average score
33.1%
Margin
±3.87
The web overall
34.3%

These do not shape the site, so the figure describes their clientele: sites that take card payments are better built, and the credit goes to the sites themselves.

CDN averages 33.1%33.1%
23
UI frameworks

UI frameworks

Technologies
18
Average score
33%
Margin
±5.15
The web overall
34.3%

These do not shape the site, so the figure describes their clientele: sites that take card payments are better built, and the credit goes to the sites themselves.

UI frameworks averages 33%33%
18
Web servers

Web servers

Technologies
13
Average score
32.9%
Margin
±4.48
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Web servers averages 32.9%32.9%
13
Hosting

Hosting

Technologies
26
Average score
32.8%
Margin
±5.15
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Hosting averages 32.8%32.8%
26
Programming languages

Programming languages

Technologies
9
Average score
32.3%
Margin
±5.27
The web overall
34.3%

These do not shape the site, so the figure describes their clientele: sites that take card payments are better built, and the credit goes to the sites themselves.

Programming languages averages 32.3%32.3%
9
Web frameworks

Web frameworks

Technologies
16
Average score
31.9%
Margin
±5.05
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Web frameworks averages 31.9%31.9%
16
Ecommerce

Ecommerce

Technologies
21
Average score
29.9%
Margin
±5.27
The web overall
34.3%

These technologies shape the site, so the figure says something about the technology.

Ecommerce averages 29.9%29.9%
21

Markets

All 107 suffixes

What a domain suffix can and cannot tell you

It says nothing about governments or universities: the suffixes that reliably identify an institution are small populations, and none of them reaches the publication threshold. Nor does a suffix give the sector: a .com can be anything, so nothing here describes a particular industry.

The average score of the sites under each domain suffix, such as .com or .org.

SuffixAWCSMonth by month
.com

.com

Sample
n=56,803
Coverage
95%
Margin
±0.22
The web overall
34.3%

Two figures closer together than ±0.22 count as a tie.

.com: 35.7%35.7%
.com: 5 months, no significant change= +0.2
.de

.de

Sample
n=5,362
Coverage
97.8%
Margin
±0.72
The web overall
34.3%

Two figures closer together than ±0.72 count as a tie.

.de: 34.8%34.8%
.de: 5 months, no significant change= +0.7
.com.br

.com.br

Sample
n=3,730
Coverage
93.4%
Margin
±0.87
The web overall
34.3%

Two figures closer together than ±0.87 count as a tie.

.com.br: 34.4%34.4%
.com.br: 5 months, up▲ +1.2
.fr

.fr

Sample
n=2,863
Coverage
95%
Margin
±0.99
The web overall
34.3%

Two figures closer together than ±0.99 count as a tie.

.fr: 33.9%33.9%
.fr: 5 months, no significant change= +0.7
.net

.net

Sample
n=3,110
Coverage
95.8%
Margin
±0.95
The web overall
34.3%

Two figures closer together than ±0.95 count as a tie.

.net: 33.8%33.8%
.net: 5 months, no significant change= +0.6
.org

.org

Sample
n=4,697
Coverage
96.7%
Margin
±0.77
The web overall
34.3%

Two figures closer together than ±0.77 count as a tie.

.org: 32.9%32.9%
.org: 5 months, up▲ +1
.it

.it

Sample
n=3,025
Coverage
96.2%
Margin
±0.96
The web overall
34.3%

Two figures closer together than ±0.96 count as a tie.

.it: 31.2%31.2%
.it: 5 months, no significant change= +0.7
.ru

.ru

Sample
n=3,381
Coverage
96.6%
Margin
±0.91
The web overall
34.3%

Two figures closer together than ±0.91 count as a tie.

.ru: 25.9%25.9%
.ru: 5 months, no significant change= −0

AI agents

All 17 agents

How many sites name each AI agent in their robots.txt.

AgentSites naming itMonth by month
GPTBot

gptbot

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

GPTBot: 9.6%9.6%
GPTBot: 5 months, no significant change= −0 pp
ClaudeBot

claudebot

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

ClaudeBot: 8.8%8.8%
ClaudeBot: 5 months, up▲ +0.2 pp
CCBot

ccbot

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

CCBot: 8.2%8.2%
CCBot: 5 months, down▼ −0.2 pp
Google-Extended

google-extended

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

Google-Extended: 7.9%7.9%
Google-Extended: 5 months, down▼ −0.1 pp
Bytespider

bytespider

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

Bytespider: 7.5%7.5%
Bytespider: 5 months, down▼ −0.2 pp
Amazonbot

amazonbot

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

Amazonbot: 7.5%7.5%
Amazonbot: 5 months, down▼ −0.3 pp
meta-externalagent

meta-externalagent

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

meta-externalagent: 7.2%7.2%
meta-externalagent: 5 months, down▼ −0.2 pp
Applebot-Extended

applebot-extended

Sample
n=145,652
Coverage
100%
Margin
±0.14 pp
The web overall
34.3%

Two figures closer together than ±0.14 pp count as a tie.

Applebot-Extended: 7.1%7.1%
Applebot-Extended: 5 months, down▼ −0.2 pp

Reports

All 5 reports

One report for each month of the census, unchanged once published so it can be cited.

MonthAWCSMonth by month
September 2026
September 2026: 34.3%34.3%
September 2026: 5 months, up▲ +0.4
August 2026
August 2026: 33.9%33.9%
August 2026: 4 months, up▲ +0.6
July 2026
July 2026: 33.3%33.3%
July 2026: 3 months, up▲ +1.6
June 2026
June 2026: 31.7%31.7%
June 2026: 2 months, up▲ +1.5
May 2026
May 2026: 30.2%30.2%
—