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.

Holistic Evaluation of Language Models (HELM) is an open source Python framework created by the Center for Research on Foundation Models (CRFM) at Stanford for holistic, reproducible and transparent evaluation of foundation models, including large language models (LLMs) and multimodal models.

01Result

stanford-crfm/helm ↗

stanford-crfm/helm AI coding statistics: 2.5%

of 161 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

2.5% agent · top contributor 75%2 contributors

individual

One person is driving this. The headline number reflects a single contributor's workflow, not a project-wide shift.

agent commits

157

other considered commits

agent contributors

75%

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

4assisted

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: 6.7%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Claude Code	Anthropic	3	1.9%
Cursor Agent	Anysphere	1	0.6%

03Evidence

Inspect the matches

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

Matched commit samples

d4ba787c on GitHubFix print_summary() showing relevance scores under COHERENCE (#4183)Claude Code
bfd9b963 on GitHubAdd link to per_instance_stats.json on frontend (#4166)Claude Code
5dede25d on GitHubFix _apply_output_mapping_pattern returning wrong match results (#4192)Claude Code
d7f74a37 on GitHubdocs: fix README typos (#4287)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	192
Merge commits removed	−0
CI / dependency bots removed	−31
Considered	161

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 →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 stanford-crfm/helm

Scan another repository

Try

Month	Agent share	Considered commits
2026-02	0%	28
2026-03	0%	39
2026-04	6.7%	45
2026-05	2.4%	41
2026-06	0%	8