Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.
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: 21.9%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 29 | 8.6% |
| GitHub Copilot | GitHub | 5 | 1.5% |
| Cursor Agent | Anysphere | 4 | 1.2% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
45fd4a04docs(security): destructive path targets need an allowlist, a depth floor,Claude Code6a9f2eb2docs(observability): name the entry point when several write to one logClaude Code1483fc12fix(skills): add database-migration and dependency-audit vocabulary to twoClaude Code75dfcb33Potential fix for pull request findingGitHub Copilot77ffedb2fix: update marketplace.json to use explicit GitHub source formatGitHub Copilot4d277aa7docs(skills): tighten frontmatter for idea-refine and browser-testingGitHub Copilotbb68af56fix(hooks): quote opt-in hook paths in setup docsCursor Agentb68f553adocs: add skill-gap issue form for reporting guidance mismatchesCursor Agentf25f467efix(validator): reject fenced-block headings and negated triggers (#387)Cursor Agent
What was excluded
| Commits in window | 512 |
| Merge commits removed | −173 |
| CI / dependency bots removed | −0 |
| Considered | 339 |
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 addyosmani/agent-skills