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: 100%. Bars are scaled to that peak, not to 100%.
Which agents
| OpenAI Codex | OpenAI | 10 | 3.7% |
| Claude Code | Anthropic | 4 | 1.5% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
e6e6cd23ci: publish versioned PUA release notes and skill packagesOpenAI Codex4a779385fix: preserve PUA tone across model runtimes in 3.5.1OpenAI Codex534b91c6fix(upload): make contribution uploads reliableOpenAI Codexa77a7f2efix(hooks): make stop-feedback.sh parse on bash 3.2 (macOS default)Claude Code7e5f7723fix(integrity-guard): skip bare identifiers as path candidatesClaude Code62f8a1a6fix: 修复 /pua:off 无法禁用沮丧检测和失败升级 hook 的问题Claude Code
What was excluded
| Commits in window | 288 |
| Merge commits removed | −21 |
| CI / dependency bots removed | −0 |
| Considered | 267 |
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 tanweai/pua