Review economics
openai/evals
Read from the 100 most recently merged pull requests ·
Pull requests with attributed agent authorship took only 1.0× the reviews submitted of the rest.
≥15%
Attributed
a floor
20
Attributed PRs
80
Other PRs
100
Read
Attributed — the latest 5
#1 Add in Reverse String eval
1 round · 2 reviews · 1h to merge
#1644 Pin pre-commit hook revisions to immutable commits
1 reviews · 21h to merge
#1637 [codex] Pin GitHub Actions workflow references
1 reviews · 11.6d to merge
#1605 Remove incontext_rl suite with defunct dependencies
1 reviews · 2h to merge
#1572 Updating readme to link to OpenAI hosted evals experience
1 reviews · 0h to merge
The rest — the latest 5
#926 Auditing & Other Assurance Services (Evals)
3 reviews · 3.1d to merge
#972 Fix get_answer
1 reviews · 33h to merge
#771 for lack of a better name, "gpt protocol buffers"
1 round · 2 reviews · 46.4d to merge
#1528 [eval] Add IMO problems with exact answers
1 round · 2 reviews · 59.6d to merge
#1560 20240930 steven exception handling usage tokens
1 reviews · 0h to merge
How this was measured
15% 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; reads 100 of 655 merged PRs — the rest have not been analyzed yet.
Detected: OpenAI Codex.
“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 openai/evals 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 openai/evals — 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