Review economics
shareAI-lab/learn-claude-code
Read from the 72 most recently merged pull requests ·
Pull requests with attributed agent authorship took 1.1× the time to merge of the rest.
≥14%
Attributed
a floor
8
Attributed PRs
64
Other PRs
72
Read
Attributed — the latest 5
#142 fix: append nag reminder instead of inserting at front
0 reviews · 9h to merge
#364 refactor: summarize only trimmed history in reactive_compact
0 reviews · 6.1d to merge
#318 fix(s03): route permission-overview deny arrow back to messages[]
0 reviews · 25h to merge
#322 fix(s06): move extract_text() next to its only caller
0 reviews · 6h to merge
#159 Fix unhandled OSError in subprocess and unsafe dict access in subagent
0 reviews · 16h to merge
The rest — the latest 5
#490 fix(s18): update wt_ctx after idle_poll auto-claims a worktree task
0 reviews · 0h to merge
#489 docs(s16): fix protocol flowchart — separate shutdown and plan approval flows
0 reviews · 0h to merge
#484 fix(s03,s04,s20): resolve Gate2 safe_path conflict — permission dead on write/edit
0 reviews · 0h to merge
#488 fix(s10): assemble system prompt from dynamic context instead of static strings
0 reviews · 0h to merge
#487 docs(s08): clarify reactive_compact aggressive wording refers to trigger, not compression
0 reviews · 0h to merge
How this was measured
14% of merged commits carry agent attribution — a floor, not the share; tools that only complete code inline leave no commit trail, so the unattributed side includes AI-assisted work; 4 PRs came off an agent branch with no commit attribution, so the real share runs higher.
Detected: Claude Code, OpenAI Codex, GitHub Copilot coding agent.
“Attributed” means a commit carried an agent’s signature — a co-author trailer, an agent commit identity, or an agent bot account. Tools that only complete code inline leave no such mark, so the other column is “rest”, not “human-written”. Full method and its limits
The badge reads this repo’s current report, so it follows the number.
Tell me when this moves
We re-read shareAI-lab/learn-claude-code weekly and email only when the number changes materially.
Read your own repos
This one is public. For a private repo, run the same read locally through your own GitHub credentials — nothing is installed and nothing is sent to us.
npx @ambera/review-taxWant this continuously on shareAI-lab/learn-claude-code — each pull request paired to the task it came from, and the work graded from its review loop rather than its diff size? Claim this repo in Forge
Browse every repo people have read · Read a different repository