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.

Intercept and inspect Coding Agent API traffic from Claude Code, Codex CLI, Gemini CLI, Cursor CLI, OpenCode, Kimi/Kimi Code, Pi, and Hermes in a local trace viewer.

01Result

liaohch3/claude-tap ↗

liaohch3/claude-tap AI coding statistics: 19.3%

of 373 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate
19.3% agent · top contributor 30.6%11 contributors
org-wide

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

72
agent commits
301
other considered commits
11
agent contributors
30.6%
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

72assisted
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
2026-0240.5%79
2026-0369%29
2026-0418.2%11
2026-055.6%126
2026-061.5%66
2026-077.1%42
2026-0836.8%19
2026-090%1

Peak month: 69%. Bars are scaled to that peak, not to 100%.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic6617.7%
Cursor AgentAnysphere92.4%

03Evidence

Inspect the matches

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

Matched commit samples

  • 8c53c1db on GitHubfeat(viewer): label harness-injected user input and title groups by human Claude Code
  • 06c95936 on GitHubfeat(viewer): explain prompt cache misses in the turn detail (#437)Claude Code
  • d87e7fe7 on GitHubfix(ci): stop redownloading the playwright browser on every run (#447)Claude Code
  • 8c53c1db on GitHubfeat(viewer): label harness-injected user input and title groups by human Cursor Agent
  • 06c95936 on GitHubfeat(viewer): explain prompt cache misses in the turn detail (#437)Cursor Agent
  • a5f68883 on GitHubfeat(viewer): move cost computation into a Python pricing adapter (#445)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 window396
Merge commits removed−7
CI / dependency bots removed−16
Considered373

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 liaohch3/claude-tap

Scan another repository

Try