Stats
Actions
Tags
From claude-workflow
Provides challenge, verification, and reset prompt patterns to push back on Claude's generic responses, prove code works, and restart suboptimal solutions. Use for rigorous code reviews and adversarial testing.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-workflow:advanced-promptingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use these when Claude's first response feels safe/generic:
Use these when Claude's first response feels safe/generic:
Force Claude to critique rather than accept:
Demand evidence over assertions:
Break out of local optima:
| Situation | Prompt Pattern |
|---|---|
| Code review feels shallow | Challenge prompts |
| Implementation seems fragile | Verification prompts |
| Solution feels hacky | Reset prompts |
| Claude is agreeing too readily | "Push back on this" |
npx claudepluginhub spences10/claude-code-toolkit --plugin claude-workflowChallenges agent's recent claims and forces genuine re-examination using code-reading tools when users push back with 'wtf are you thinking', 'that's wrong', or similar.
Analyzes user prompts for clarity, scope, specificity, success criteria, and length; provides S/A/B/C/D/F scores and concrete suggestions. Rewrites prompts on request for better Claude Code results.
Applies adversarial fresh-context review to non-trivial decisions in code. Use when correctness matters more than speed, in unfamiliar code, or for high-stakes operations.