Dataset: 2026-10-27-catalog-ai-v2
Ruleset: b2e8889c66f7
Scanned: 2026-10-27
Window: February–September 15, 2026

Catalog tracking observation. This repository is measured with the same fixed window and ruleset, but it is not part of the published benchmark or its aggregate claims.

ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

01Result

wanshuiyin/Auto-claude-code-research-in-sleep ↗

wanshuiyin/Auto-claude-code-research-in-sleep AI coding statistics: 74.1%

of 1,081 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

74.1% agent · top contributor 19.6%44 contributors

org-wide

Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.

801

agent commits

280

other considered commits

agent contributors

19.6%

largest contributor

02Drivers

What shapes this result

Mode, timing, and detected signatures show whether the result reflects broad adoption or one concentrated workflow.

Assisted or autonomous

801assisted

0autonomous

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

Peak month: 90.5%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Claude Code	Anthropic	791	73.2%
GitHub Copilot	GitHub	7	0.6%
OpenAI Codex	OpenAI	3	0.3%

03Evidence

Inspect the matches

Public commit samples let you inspect the signatures behind the aggregate result.

Matched commit samples

d9adbe0c on GitHubdocs(research-implement-feature): drop the unsupported assume: override; aClaude Code
151bc19a on GitHubfeat(skill): /research-implement-feature — build from "implement X" with aClaude Code
15b9b27d on GitHubdocs: OrcaReplay row to house length, note what the trace records, CN row Claude Code
420c5c1a on GitHubApply suggestion from @CopilotGitHub Copilot
a740c367 on GitHubApply suggestion from @CopilotGitHub Copilot
07a7e29c on GitHubApply suggestion from @CopilotGitHub Copilot
5725459e on GitHubdocs(readme): cite AI review tools and caveat scoresOpenAI Codex
e1a44884 on GitHubdocs(readme): emphasize community submissionsOpenAI Codex
34d5a12e on GitHubdocs(readme): clarify community submission review statusOpenAI Codex

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	1,182
Merge commits removed	−101
CI / dependency bots removed	−0
Considered	1,081

Bots are excluded from the denominator entirely. Agent attribution requires a literal supported signature; no code-style inference is used.

Continue exploring

Claude Code commitsBrowse repositories with detected Claude Code signatures →GitHub Copilot commitsBrowse repositories with detected GitHub Copilot signatures →OpenAI Codex commitsBrowse repositories with detected OpenAI Codex signatures →Python AI coding statisticsCompare evidence across Python repositories →GitHub AI commit analyzerInspect another repository with the same evidence model →

Reproduce this observation

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 wanshuiyin/Auto-claude-code-research-in-sleep

Scan another repository

Try

Month	Agent share	Considered commits
2026-03	85%	426
2026-04	65.4%	159
2026-05	67.3%	223
2026-06	88.1%	84
2026-07	67.7%	99
2026-08	36.2%	69
2026-09	90.5%	21