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 open-source AI workbench for scientific research

01Result

synthetic-sciences/openscience ↗

synthetic-sciences/openscience AI coding statistics: 8.5%

of 774 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

8.5% agent · top contributor 60.6%3 contributors

individual

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

agent commits

708

other considered commits

agent contributors

60.6%

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

66assisted

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: 34.7%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Cursor Agent	Anysphere	44	5.7%
Claude Code	Anthropic	22	2.8%

03Evidence

Inspect the matches

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

Matched commit samples

2956c3ba on GitHubQA pass on the autoresearch study: grants that arrive keep processes aliveCursor Agent
adcfd767 on GitHubcandidate-registry: hash tarballs through a node stream for the backend tyCursor Agent
533d74d3 on GitHubRehearsal tests follow the candidate-registry flow; test the registry end Cursor Agent
d56d62e0 on GitHubName the benchmarks correctly and link them from the figures (#556)Claude Code
249161c6 on GitHubPublish the Windows installer unsigned while any Artifact Signing value isClaude Code
5a9ac034 on GitHubPublish the Windows installer unsigned until Artifact Signing is configureClaude 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	854
Merge commits removed	−50
CI / dependency bots removed	−30
Considered	774

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 →TypeScript AI coding statisticsCompare evidence across TypeScript 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 synthetic-sciences/openscience

Scan another repository

Try

Month	Agent share	Considered commits
2026-07	2.2%	357
2026-08	0%	250
2026-09	34.7%	167