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: 80.3%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 879 | 61.8% |
| Cursor Agent | Anysphere | 77 | 5.4% |
| GitHub Copilot | GitHub | 1 | 0.1% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
40be934afix(telegram): wrap-ups on real session end from the final summary, one shClaude Codec61f8be5feat(telegram): session wrap-up notifications from the Stop summary (plan Claude Code07ba05aefix(skills): correct the timeline-report example SQL schema (#3407)Claude Codeed57a511chore: bump version to 13.24.23 (#4056)Cursor Agent91ad9ddefix(supervisor): jail SDK subprocess cwd (#4054)Cursor Agent7186eb56chore: bump version to 13.24.21 (#4052)Cursor Agent8ec91e7ffix: break infinite summary-retry loop (#1633) (#2072)GitHub Copilot
What was excluded
| Commits in window | 1,643 |
| Merge commits removed | −205 |
| CI / dependency bots removed | −15 |
| Considered | 1,423 |
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 thedotmack/claude-mem