Skills
Your AI code-review bot needs inline code suggestions — it's the single strongest predictor of whether a comment gets acted on
A study of 54,791 AI-agent code review comments across five agents (arXiv 2607.21997, July 24) found the presence of an inline code suggestion was the strongest predictor of resolution, while lengthy, complex comments were significantly less likely to be acted upon. Resolution rates varied widely by agent, with Copilot at 72.9%. Qualitative coding of 470 unresolved discussions produced ten dismissal patterns led by incorrect suggestions and intentional design decisions. Concrete config changes: make your review bot emit patches rather than prose, cap comment length, and route design/evolvability comments to core maintainers since peripheral contributors mostly engage with functional-defect comments.
↳ Follow the thread