- Dataset
- 2026-10-27-catalog-ai-v2
- Ruleset
- b2e8889c66f7
- Scanned
- 2026-10-27
- Window
- February–September 15, 2026
Open memory & reliability layer for AI agents + independent judging/verification (NACRE judge, caliber-bench meta-benchmark, preregistered criteria) — by Nautilus Platform
01Result
chunxiaoxx/nautilus-compass ↗chunxiaoxx/nautilus-compass AI coding statistics: 15.2%
One person is driving this. The headline number reflects a single contributor's workflow, not a project-wide shift.
02Drivers
What shapes this result
Mode, timing, and detected signatures show whether the result reflects broad adoption or one concentrated workflow.
Assisted or autonomous
Assisted — a developer’s identity authored the commit and the agent added itself as co-author. Autonomous — a bot identity authored the commit without a human identity attached.
Agent share by month
| Month | Agent share | Considered commits |
|---|---|---|
| 2026-04 | 40% | 25 |
| 2026-05 | 2.3% | 175 |
| 2026-06 | 2.1% | 95 |
| 2026-07 | 23.9% | 138 |
| 2026-08 | 24.8% | 226 |
| 2026-09 | 13.4% | 269 |
Peak month: 40%. Bars are scaled to that peak, not to 100%.
Which agents
| Agent | Vendor | Commits | Share |
|---|---|---|---|
| Claude Code | Anthropic | 141 | 15.2% |
03Evidence
Inspect the matches
Public commit samples let you inspect the signatures behind the aggregate result.
Matched commit samples
3fe4b5faon GitHubdocs(plans): C 族首批校准预注册判据冻结——V1-V4(可复算性/三态实弹/双向记账/签名回执)+空白区零占位句(护栏2)+FAIL Claude Codef9a34d14on GitHubfix(catalog): Updates 段归位收尾(格式纪律)Claude Code92525119on GitHubfeat(catalog): anchor-pool-selection-bias-v1 立条——flywheel 闸③三组对照归因互锚(锚=历史合Claude Code
04Method
What entered the denominator
The result begins with commit history, removes merges and routine automation, and reports literal agent signatures as a lower bound.
Exclusion ledger
| Commits in window | 943 |
| Merge commits removed | −12 |
| CI / dependency bots removed | −3 |
| Considered | 928 |
Bots are excluded from the denominator entirely. Agent attribution requires a literal supported signature; no code-style inference is used.
Continue exploring
Related coding-agent evidence
Run the same detector locally.
Deterministic signature matching; no code-style inference. Squash merges can remove trailers, so this result remains a measured floor.
npx qmmit-cli chunxiaoxx/nautilus-compass