Kraishan followed 37,623 provenance-labelled pull requests from OpenAI Codex, Devin, GitHub Copilot, Cursor and Claude Code across 2,807 GitHub repositories between December 2024 and July 2025, alongside 4,027 human pull requests drawn from…
In the News: September 20, 2026 (Extra)
Across 37,623 labelled pull requests, Codex changes were reverted at half the human rate and Devin at 1.3 times it. Agents differ most from each other.
Extra edition
Machine-readable
Download Markdown
Story
Revert rates for agent pull requests split by vendor, not by whether a human wrote the code
Read story →
Story
A practitioner names manual testing as what caps his speedup
Khramtsov sets out seven defensive layers he has seen hold defect rates flat while output rises. Six of them are ordinary quality practice with agents attached: requirements review, unit tests at high coverage, manual testing, end-to-end…