signalHacker News Best2026-10-04
Getting the most out of Opus 5.5 in Claude and Claude Code
Opus 5.5 in Claude and Claude Code excels at long, multi-step coding tasks, running for hours with minimal oversight. It always thinks before replying, so you should remove explicit 'think carefully' prompts and instead give a complete task with a defined finish line. For design work, list specific patterns to avoid, and in Claude Code, add stop/continue rules to CLAUDE.md and type follow-ups mid-run.
- for who
- Developers and Claude users working with Opus 5.5 in Claude apps and Claude Code.
- why now
- Opus 5.5's longer autonomous runs and new prompt guidance demand updated workflows immediately.
- what changes
- Prompting shifts from step-by-step instructions to whole-task briefs with clear finish lines, and from explicit thinking prompts to relying on the model's built-in reasoning.
- to do
- Update your prompts and CLAUDE.md to give the full task in one message, name the finish line, remove 'think carefully' lines, and add explicit stop/continue rules.
key points
- Opus 5.5 excels at long, multi-step coding runs that can last hours
- Always thinks before replying, so drop explicit 'think carefully' lines
- Name the finish line and list specific design patterns to avoid
#Claude Opus 5.5#prompt engineering#Claude Code
score
score 8 out of 10. 0-10: how dense the facts are, multiplied by how much you can do with them after reading. 8+ means the topic's evidence bar is met: benchmarks and availability for a new model, amount and investors for a funding round, revenue figures for a solo-money story. Below 5 an item does not enter the digest. A press release scores 3 or less, a reprint loses 2, anything older than 14 days loses 1, a headline that misleads loses 3.
read the source