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: 76%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 386 | 27.4% |
| Cursor Agent | Anysphere | 126 | 8.9% |
| GitHub Copilot | GitHub | 72 | 5.1% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
a8736a07fix(lbug): checkpoint race in pool-adapter.ts + pin @ladybugdb/core to 0.1Claude Codeceaff27cfix(parse-cache): retire a chunk whose durable generation could not be resClaude Code7bbaf6b7fix(review): fail closed on foreign embedding identity and vector-width drClaude Codeac9a4e9afix(embeddings): keep ONNX off the install and analyze critical path (#328Cursor Agent21a52af1fix(lbug): ship FTS per-platform and recover in-place native aborts (#3274Cursor Agent79543c8ffeat(storage): add configurable index storage and content retention tiers Cursor Agent12763a40fix(watch): await a watcher re-arm barrier so gitignore reloads cannot droGitHub Copilotd3d4fa31fix(scope-resolution): gate C#/Kotlin free calls by instance ownership (#2GitHub Copilot450cebc2fix(java): JLS binary-name identities for local classes, enums, records & GitHub Copilot
What was excluded
| Commits in window | 1,964 |
| Merge commits removed | −205 |
| CI / dependency bots removed | −348 |
| Considered | 1,411 |
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 abhigyanpatwari/GitNexus