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: 38.7%. Bars are scaled to that peak, not to 100%.
Which agents
| Claude Code | Anthropic | 9 | 4.3% |
| Cursor Agent | Anysphere | 9 | 4.3% |
| GitHub Copilot | GitHub | 9 | 4.3% |
| OpenAI Codex | OpenAI | 2 | 1% |
Verify it yourself
Sample commits that matched. Every one links to GitHub — check the trailers.
2b183ab3fix(esp-vocat): apply V1.2 pins on the cached path tooClaude Code213f33bafix(esp-vocat): stabilize PCB version detection across soft resetsClaude Coded68505f5Add pt-BR (Brazilian Portuguese) locale (#2243)Claude Codeac6deed3Disable C++ exceptions and keep ESP32-S31 firmware in the 4MB OTA slot (#2Cursor Agent7aa92666fix: keep AFE and LVGL out of the internal SRAM bottleneckCursor Agent3cd9385ffix(nt26): handle uart-eth-modem 0.6 events and QueryCellInfoCursor Agentd88d2c89fix(esp-hi): fit 4MB assets and restore ST7735 bring-up (#2238)GitHub Copilot01212107Fix mirrored camera image on AtomS3R-M12 (OV3660) (#2155)GitHub Copilot49ac8a6dchore: upgrade version to 2.2.6 (#1944)GitHub Copilot2a9157dbBind Otto MCP responses to request client (#2021)OpenAI Codexe182471fValidate MQTT goodbye session id (#2023)OpenAI Codex
What was excluded
| Commits in window | 241 |
| Merge commits removed | −31 |
| CI / dependency bots removed | −0 |
| Considered | 210 |
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 78/xiaozhi-esp32