76.8%

of 1,192 commits were committed by an AI agent — a floor, not an estimate

76.8% agent · top contributor 75.7%37 contributors

team

A small group is using agents. Adoption has spread beyond one person but is not project-wide.

916

agent commits

276

human commits

agent contributors

75.7%

largest contributor

Assisted or autonomous

193assisted

723autonomous

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

Jan26

Feb26

Mar26

Apr26

May26

Jun26

Jul26

Aug26

Sep26

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

Which agents

Devin	Cognition	719	60.3%
Claude Code	Anthropic	201	16.9%
GitHub Copilot	GitHub	2	0.2%
Cursor Agent	Anysphere	1	0.1%

Verify it yourself

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

d695f41bFix date-dependent failure in interval schedule form test (#23125)Devin
253a8fc9ui-v2: improve dark-mode readability of flow run cards (#23114)Devin
09fe6a7eFormat deployment Created/Updated timestamps in ui-v2 metadata (#23112)Devin
c3496c2aAdd Redis Sentinel support to prefect-redis (#22377)Claude Code
1efe2981ci: pass the release ref through env in the prerelease tag check (#22882)Claude Code
517a7c41Confirm suspension events against the flow run instead of state-by-id (#22Claude Code
467c391ffix: apprise-logger-level-normalization (#21709)GitHub Copilot
c62da6b6fix: correct return type annotation for dbt_runner_freshness_success fixtuGitHub Copilot
9d0940affix(ui-v2): allow any float precision on schema number inputs (#22643)Cursor Agent

What was excluded

Commits in window	1,601
Merge commits removed	−0
CI / dependency bots removed	−409
Considered	1,192

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 PrefectHQ/prefect

Scan another repository

Try