Crank up reasoning with /effort high
Claude has been under-allocating thinking on hard tasks — `/effort high` (or `max` on Opus) plus `showThinkingSummaries` brings the deep analysis back.
- Type: Prompt
- Tags: Coding Agent Hacks, Intermediate, Claude Code
- Updated: 2026-10-01
Prompt
Claude has been under-allocating thinking on hard tasks — `/effort high` (or `max` on Opus) plus `showThinkingSummaries` brings the deep analysis back. After analyzing 17,000+ thinking blocks, an AMD engineer found Claude was skipping ahead and editing code before fully processing the problem. PM Paweł Huryn's three fixes: run `/effort high`, set `showThinkingSummaries: true`, and pin a CLAUDE.md rule against editing unread code. For nuclear control, set `CLAUDE_CODE_DISABLE_ADAPTIVE_THINKING=1` in env. Implementation example: # In Claude Code: /effort high # any model /effort max # Opus, hard debugging # ~/.claude/settings.json { "showThinkingSummaries": true } # CLAUDE.md "Research the codebase before editing. Never change code you haven't read."
More prompts
- Expense Report Automation System — Save More Time
- Meeting Scheduling Link Standardization — Save More Time
- Invoice Generation and Delivery Automation — Save More Time
- Document Template Automation Library — Save More Time
- Calendar Management Automation Rules — Save More Time
- File Organization and Naming System — Save More Time
- Recurring Payment Automation Setup — Save More Time
- Data Entry Elimination Framework — Save More Time