One person is driving this. The headline number reflects a single contributor's workflow, not a project-wide shift.
Assisted or autonomous
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
Peak month: 86%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 841 | 63.6% |
| Cursor Agent | Anysphere | 75 | 5.7% |
| GitHub Copilot | GitHub | 1 | 0.1% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
00095adbFix: skipScan must cover the visual contrast stage tooClaude Code46f13989Fix: write the build-path flip before answering the POSTClaude Code152d6940Fix: harden live overlay detector waivers (#639 follow-up)Claude Code8b326fc8Fix: resolve unique --target names in monorepos (#700) (#706)Cursor Agent38e102f0Fix: never inject raw JSX in live-mode fallback (#454) (#694)Cursor Agent377fb112Pass the session key from detached idle-grace testsCursor Agent36457e19Bump skill-behavior test lineup: gemini-3.7-flash (#598)GitHub Copilot
What was excluded
| Commits in window | 1,686 |
| Merge commits removed | −126 |
| CI / dependency bots removed | −238 |
| Considered | 1,322 |
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 pbakaus/impeccable