Agentic Web Census

August 2026by Omitsis

English

Census report · May 2026

The first month of the census finds the web 14.3% ready for AI agents, with nothing at all exposing tools an agent could call.

This is May 2026, frozen. It is the first month of the series, so there is nothing to compare it against yet.

What moved it

Poor14.3%

AWCS

What the AWCS is, and how it is measured

Measured on 14,577 pages, 95.9% of which carried the agentic category, with a confidence margin of ±0.44.

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 notApplicable 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=14,374 · coverage 95.8% · margin ±0.44 pp.

53% of the score
Read on its full scale

Weak30.4%

Read: 4 months, up▲ +1.3 pp
Guide

What Guide measures

The llms-txt audit passing: a file is served 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 index uses, because a broken file does not guide an agent.

n=14,373 · coverage 95.8% · margin ±0.44 pp.

27% of the score
Guide on its full scale

Poor13.1%

Guide: 4 months, no significant change= +0.1 pp
Act

What Act measures

A count, not a percentage: how many sampled sites register WebMCP tools that an agent can call. It is the audit that defines the Act dimension — the other two WebMCP audits measure how good those tools are, not whether they exist.

n=14,374 · coverage 95.8% · margin ±0.44 pp.

20% of the score
Act on its full scale

Very poor5.1%

Act: 4 months, up▲ +0.6 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. Gradual and not a yes or no — turning away one agent out of seventeen is not the same as turning away all of them. A site with no robots.txt blocks nobody, so it counts as fully reachable: that is a measurement, not a gap. This is the only figure that MULTIPLIES the index 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 we would rather state than hide: a blanket Disallow under User-agent: * is not counted, because the corpus-wide measurement cannot see it either, and we would rather be consistent than clever.

n=15,009 · coverage 100.0% · margin ±0.43 pp.

multiplies the rest
Reachable on its full scale

Excellent96.0%

Reachable: 4 months, no significant change= −0.2 pp

Every figure of the month

Each one with the sample it was measured on and the margin that comes with it.

FigureMay 2026PagesMargin
AWCS14.3%14,577±0.44
Pages an agent can parse26.2%14,577±0.44 pp
Agents that can reach the site96.4%15,208±0.43 pp
Sites with a valid llms.txt3.1%14,578±0.44 pp
Sites exposing tools014,579±64 sites
Sites serving an llms.txt29.9%14,578±0.44 pp
Sites with an AI crawler policy8.9%15,208±0.43 pp
Layout stability87.1%14,579±0.44
Google's category score52.2%14,576±0.44

The two ends

The five highest and the five lowest technologies of the month, by index.

EndTechnologyAWCSPages
HighestHydrogen80.0%1
HighestFirstPromoter80.0%1
HighestChili Piper80.0%1
HighestaThemes Astrid53.3%1
HighestPlanet53.3%1
LowestShoplazza0.0%3
LowestXtra0.0%2
LowestLivePerson0.0%4
LowestHello Bar0.0%7
LowestFlask0.0%1

The data behind this report

How this report is written, and which parts a model may touch

figures passed to the model: 16
numbers it may write: only those, or their roundings
if it writes any other: the build FAILS and nothing is published
the contractThe rule is executable and not a promise: lib/prose.ts rejects a number nobody passed in, and thirteen tests hold it there.
Everything on this page is calculated — the figures, the deltas, the tables, the records and this note. The headline, the "what moved it" and the advice were written by claude-opus-5 on 2026-08-23, from the figures above and nothing else: no database access, nothing to compute. The text is stored in the repository and validated again at build time, so a number nobody passed in stops the build instead of reaching this page.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 are not one above the other, they are tied. That is why nothing here has a position number.

Every figure above, as a file. A report you cannot check is an opinion.

curl -O https://agenticwebcensus.drupal.omitsisdev.com/report/2026-05.json
shellThe aggregate of May 2026, exactly as the site read it.