---
title: "Comment by Skillselion on \"Claude Opus 5 Honest Review: \nIs It Worth Switching From Opus 4.8?\" | Hashnode"
description: "Deleting the \"double-check your answer\" lines is the tip I would act on first. Those prompts were written for models that under-verified; on a model that alread"
url: "https://hashnode.com/posts/claude-opus-5-honest-review-is-it-worth-switching-from-opus-4-8/6a69ee475164f278edca1a04/comment/6a856b1d52d3e2956ccde478"
---

[Discussion](https://hashnode.com/posts/claude-opus-5-honest-review-is-it-worth-switching-from-opus-4-8/6a69ee475164f278edca1a04)

Comment [Claude Opus 5 Honest Review: Is It Worth Switching From Opus 4.8?](https://hashnode.com/posts/claude-opus-5-honest-review-is-it-worth-switching-from-opus-4-8/6a69ee475164f278edca1a04)

[Skillselion](https://hashnode.com/@skillselion)

[Aug 19](https://hashnode.com/posts/claude-opus-5-honest-review-is-it-worth-switching-from-opus-4-8/6a69ee475164f278edca1a04/comment/6a856b1d52d3e2956ccde478)

Deleting the "double-check your answer" lines is the tip I would act on first. Those prompts were written for models that under-verified; on a model that already self-verifies they read as an instruction to keep going, which is where a chunk of the over-engineering you describe comes from. Same experience on subagent sprawl: capping how many it may spawn changed our bill more than any effort-level setting did. One place I would soften the verdict: 55.2% vs 61.1% recall makes it sound unfit for code review, but a precision-heavy reviewer works well as the second voice in an ensemble where a cheaper high-recall pass goes first. As the only safety net, agreed, no.