Pulse
2026-09-18-v3.0
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.

Terminal based IDE & text editor: easy, powerful and fast

01Result

sinelaw/fresh ↗

sinelaw/fresh AI coding statistics: 91.3%

of 5,580 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate
91.3% agent · top contributor 86.4%16 contributors
individual

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

5,096
agent commits
484
other considered commits
16
agent contributors
86.4%
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

5096assisted
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
MonthAgent shareConsidered commits
2026-0282.7%456
2026-0391%576
2026-0493.5%1207
2026-0590.6%1115
2026-0687.3%810
2026-0796.1%684
2026-0894.2%521
2026-0995.7%211

Peak month: 96.1%. Bars are scaled to that peak, not to 100%.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic508891.2%
Cursor AgentAnysphere60.1%
GitHub CopilotGitHub20%

03Evidence

Inspect the matches

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

Matched commit samples

  • 1786c264 on GitHubkeys: don't read Shift off a character that has no lowercase (#3302)Claude Code
  • 600c0ef1 on GitHubFix the SSH-workspace-deletion panic, and make its shape a compile error (Claude Code
  • da558ddc on GitHubClosing a modal does not hand the keyboard back to what it blurred (#3275)Claude Code
  • 99d58eea on GitHubfix: stabilize flaky watch_path e2e test on CICursor Agent
  • 42d80d00 on GitHubfix: copy plugin lib/ directory in git_explorer test setupCursor Agent
  • 219e67fa on GitHubrefactor: unify sub-repo discovery, watch all git indexes in monorepoCursor Agent
  • 86492d20 on GitHubAdd openSUSE install instructionGitHub Copilot
  • 439c2a4b on GitHubAdded path:line[:col] parsing for Open File prompt and Quick Open, and jumGitHub Copilot

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 window5,593
Merge commits removed−1
CI / dependency bots removed−12
Considered5,580

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 →Rust AI coding statisticsCompare evidence across Rust 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 sinelaw/fresh

Scan another repository

Try