Users say Claude Opus 5 is strong for worlds and simulations but breaks coding workflows
omarsar0 · x · 2026-07-29
The thread says Claude Opus 5 is surprisingly strong at building complex worlds, animations, and simulations, but still lags behind GPT-5.6-Sol as a coding agent.
A follow-up reply is much harsher: after a few sessions, the user stopped using Opus 5 because it felt “ignorant” — it did things that were never asked for, ignored custom skills and tools, and broke existing workflows and loops. The core complaint is that the model is not plug-and-play for agentic coding even if it can be impressive in other creative tasks.
Related event: Claude Opus 5 Criticized: Unstable as Coding Agent, Breaks Workflows(5 posts)→
More from coding & agent
- Coding agent runs are getting long — one dev suggests breathing, not parallel sessions — teropa · 2026-10-16
- Dev builds MCP layer so Claude Code can work on servers without ever holding SSH keys — NETMESAFE · 2026-10-16
- As coding agents run longer autonomously, one dev's tip: don't parallel-session, just breathe — teropa · 2026-10-16
- 'Refusing to code with AI in 2026 is like refusing Google in 2005' — prasenx · 2026-10-16
- Claude Code adds idle compaction to save 1h prompt cache on idle sessions — EricBuess · 2026-10-16
- Handling agents that crash mid-task: the hard part is the unknown outcome across the network — Responsible-Dare5139 · 2026-10-16