Forms an agent can fill across the web

Sites whose forms are described well enough for an agent to complete. It has not measurably changed since July 2026: the gap between the two is inside the margin.

Excellent98.6%

Forms an agent can fill in September 2026

What the Forms an agent can fill in September 2026 is, and how it is measured

Measured on 7,237 pages, 4.4% of which carried the agentic category, with a confidence margin of ±0.62 pp.

= +0.2 pp

98.7%98.6%

Forms an agent can fill, July 2026 to September 2026, one measurement a month. The last step was inside the margin, so it is a tie.

What this measures

Lighthouse audit: webmcp-form-coverage

Of the sites that register WebMCP tools, those whose forms are described well enough for an agent to complete. Its mode is informative and not binary, which means Google weights it at zero and the category score does not see it.

Who moved

How to read these figures

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.

How many actually moved

0 of the 1 segments measured on both months moved by more than their margin. The other 1 did not move: their difference is smaller than this sample can resolve. All of them are in the table below.

Nobody moved

Every one of the 1 segments stayed inside its margin. The figure for the web overall can still move while none of its parts do — the corpus is measured on far more pages than any single segment, so it resolves smaller changes.

From Omitsis, who publish this census

See where your site sits on this figure

We measure it against this census and tell you what to change first.

Leave it to us