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.

The Microsoft community Windows Package Manager manifest repository

01Result

microsoft/winget-pkgs ↗

microsoft/winget-pkgs AI coding statistics: 0.3%

of 83,872 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

0.3% agent · top contributor 10%118 contributors

org-wide

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

240

agent commits

83,632

other considered commits

118

agent contributors

10%

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

240assisted

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

Agent-attributed commit share by month
Month	Agent share	Considered commits
2026-01	2.5%	40
2026-02	0.2%	7627
2026-03	0.4%	8575
2026-04	0.3%	11863
2026-05	0.2%	11882
2026-06	0.2%	11747
2026-07	0.2%	12589
2026-08	0.3%	12754
2026-09	0.7%	6795

Peak month: 2.5%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
GitHub Copilot	GitHub	140	0.2%
Claude Code	Anthropic	87	0.1%
Cursor Agent	Anysphere	10	0%
OpenAI Codex	OpenAI	3	0%

03Evidence

Inspect the matches

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

Matched commit samples

039a3ea8 on GitHubAdd shizgnit.tachikoma 0.1.0 (#418737)GitHub Copilot
404a3edd on GitHubUpdate: DavidObando.Oahu to 1.1.22 (#433458)GitHub Copilot
62b0fcbf on GitHubFix Manifest Validation Diagnosis comment permissions (#433981)GitHub Copilot
9c0f0951 on GitHubNew version: Pithflow.Pithflow version 1.37.0 (#433936)Claude Code
d73db63c on GitHubModify: DeskRipple.DeskRipple version 1.1.0 (#433914)Claude Code
8b3fb034 on GitHubRobertasTa.TempCleaner version 1.2 (#433792)Claude Code
1871108e on GitHubNew package: randlee.wyvern version 0.6.0 (#425526)Cursor Agent
b0005cb5 on GitHubLegacyFileConverter.LegacyFileConverter version 6.1.1.0 (#420067)Cursor Agent
b8d6f23f on GitHubNew package: DevCentr.Ibex version 0.2.0 (#413489)Cursor Agent
3b852e04 on GitHubUpdate: Karuzip.Karuzip to 2.2.1 (#429934)OpenAI Codex
20077f00 on GitHubNew version: Socoliuc.FollowFrame version 0.3.9 (#418658)OpenAI Codex
f024576d on GitHubNew package: Karuzip.Karuzip version 1.4.5 (#410210)OpenAI 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	83,970
Merge commits removed	−0
CI / dependency bots removed	−98
Considered	83,872

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 →Cursor Agent commitsBrowse repositories with detected Cursor Agent signatures →OpenAI Codex commitsBrowse repositories with detected OpenAI Codex signatures →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 microsoft/winget-pkgs

Scan another repository

Try