18.3%

of 109 commits were committed by an AI agent — a floor, not an estimate

18.3% agent · top contributor 90%2 contributors

individual

One person is driving this. The headline number reflects a single contributor's workflow, not a project-wide shift.

agent commits

human commits

agent contributors

90%

largest contributor

Assisted or autonomous

20assisted

0autonomous

Assisted — a developer’s identity authored the commit and the agent added itself as co-author. The agent ran git commit during a working session, with a person there. Autonomous — a bot identity is the author, no human in the loop. Across the whole index, 93.7% of agent commits are assisted.

Agent share by month

Feb26

Mar26

May26

Jun26

Jul26

Aug26

Peak month: 60.7%. Bars are scaled to that peak, not to 100%.

Which agents

Claude Code	Anthropic	20	18.3%
Cursor Agent	Anysphere	1	0.9%

Verify it yourself

Sample commits that matched. Every one links to GitHub — check the trailers.

16c12eddHoist _children into a local in constructNewChildrenArrayClaude Code
d8da70a1Fix perf regression from the attribute/unmount loop golfingClaude Code
eed9e880Golf: use Array#some for the stale-children unmount loopClaude Code
2f904c7ffix: vnode clone compatible with hardenedjs envCursor Agent

What was excluded

Commits in window	177
Merge commits removed	−68
CI / dependency bots removed	−0
Considered	109

Bots are excluded from the denominator entirely. Counting dependabot as “AI-written code” is the fastest way to produce a meaningless number.

How this was measured

Measured over February–August 2026, cloned directly. Detection is deterministic string matching against signatures agents write about themselves — no heuristics, no style analysis. Merge commits and CI bots are excluded from the denominator. The figure is a floor: squash merges strip trailers, so real usage is higher. Contributor counts are aggregate — no individual is named.

Run it yourself: npx qmmit-cli preactjs/preact

Scan another repository

Try