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.

[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

01Result

xlang-ai/OSWorld ↗

xlang-ai/OSWorld AI coding statistics: 18.5%

of 108 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

18.5% agent · top contributor 20%13 contributors

org-wide

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

agent commits

other considered commits

agent contributors

20%

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

20assisted

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: 28.6%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Claude Code	Anthropic	18	16.7%
OpenAI Codex	OpenAI	1	0.9%
Cursor Agent	Anysphere	1	0.9%

03Evidence

Inspect the matches

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

Matched commit samples

972cb73e on GitHubfix(tasks): stop 0c825995 accumulating an `environment_policy` folderClaude Code
d7d6459e on GitHubfix(tasks): stop 22a4636f accumulating a `meetings` folder per resetClaude Code
690fb857 on GitHubfix(evaluator): compare VS Code extension ids case-insensitivelyClaude Code
f8af6edd on GitHubAdd Qwen computer-use agent (#522)OpenAI Codex
51a35b73 on GitHubAdd VM secret file injection support (#503)Cursor Agent

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	112
Merge commits removed	−4
CI / dependency bots removed	−0
Considered	108

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 →Cursor Agent commitsBrowse repositories with detected Cursor Agent 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 xlang-ai/OSWorld

Scan another repository

Try

Month	Agent share	Considered commits
2026-02	0%	16
2026-03	20.8%	24
2026-04	20%	10
2026-05	23.1%	13
2026-06	28.6%	7
2026-07	26.3%	19
2026-08	17.6%	17
2026-09	0%	2