A small group is using agents. Adoption has spread beyond one person but is not project-wide.
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: 23.8%. Bars are scaled to that peak, not to 100%.
Which agents
| Cursor Agent | Anysphere | 66 | 4.5% |
| GitHub Copilot | GitHub | 45 | 3.1% |
| Claude Code | Anthropic | 40 | 2.7% |
| OpenAI Codex | OpenAI | 3 | 0.2% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
0419bb8ffeat(table): `show only` columns from the column menu and explorer (#10768Cursor Agent1ee282bdfix: map geoarrow to geoarrow-types (#10659)Cursor Agentad02456ffix(tests): compare Pygments HTML output semantically across 2.19–2.21 (#1Cursor Agentbb3532acfix: preserve unqualified references in methods (#10679)GitHub Copilot3a1b1dabfeat: lint rule for private imports (#10580)GitHub Copilotf1b3678afix: slurm script regression and docs/examples update (#10469)GitHub Copilot5aa823f9fix(markdown): render display math nested in list item continuations (#107Claude Codeb4181abafix(vega): use ctrl for chart pan/zoom on non-mac platforms (#3812) (#1049Claude Code3955fdf9fix(matplotlib): derive canvas draw scale from the backing store (#10626)Claude Codeb8c9ce29feature: Infer file browser initial path from initial value (#10362)OpenAI Codex6faa194dfeat: support default file browser selection (#10255)OpenAI Codex3b2f21a5bug: formatting of exposed classes preserves class decorators (#10042)OpenAI Codex
What was excluded
| Commits in window | 1,530 |
| Merge commits removed | −0 |
| CI / dependency bots removed | −61 |
| Considered | 1,469 |
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 marimo-team/marimo