Claude Opus 4.8 felt stricter at code review for a few hours
in Claude Code usually gave a fairly agreeable review style, often accepting work as fine. A separate was used to double-check the coding output, and it usually also approved the work unless it was asked to search online first. During a recent roughly five-hour window, the separate review seemed much better at catching flaws without online search.
It appeared to spot weak work or false approval almost immediately, with the success rate feeling close to 95%. It is still unclear whether this was a real model behavior change or just a personal impression.
Key points
- was being used inside Claude Code for project work.
- A second was used as a reviewer for the coding output.
- The review suddenly seemed better at finding flaws during a roughly five-hour period.
- The stronger review behavior appeared even without online search.
- The observation needs controlled repeat before drawing a firm conclusion.