Test Any Prompt
Universal prompt testing methodology using RED-GREEN-REFACTOR cycle with subagents, supporting A/B testing and regression testing for commands, hooks, skills, and production prompts
What is it?
Testing prompts is TDD applied to LLM instructions.
Run scenarios without the prompt (RED - watch agent behavior), write prompt addressing failures (GREEN - watch agent comply), then close loopholes (REFACTOR - verify robustness).
Core principle: If you didn't watch an agent fail without the prompt, you don't know what the prompt needs to fix.
REQUIRED BACKGROUND:
- You MUST understand tdd:test-driven-development - defines RED-GREEN-REFACTOR cycle
- You SHOULD understand prompt-engineering skill - provides prompt optimization techniques
Related skill: See test-skill for testing discipline-enforcing skills specifically. This command covers ALL prompts.
How to use it?
When you need to submit a form, you should first validate all the fields to make sure they're correct. After validation succeeds, you can proceed to submit. If validation fails, show errors to the user.
**After (37% fewer tokens):**
```markdown
Key Features
- Universal prompt testing methodology using RED-GREEN-REFACTOR cycle with subagents, supporting A/B testing and regression testing for commands, hooks, skills, and production prompts
- Seamless integration with Claude's development workflow
- Comprehensive guidelines and best practices for test any prompt
Related Skills
More from AI & MLTest Skill with Subagents
Test any Claude skill using RED-GREEN-REFACTOR cycle with subagent pressure testing to verify the skill resists agent rationalization and bypass attempts
Review Local Changes
Multi-agent code review system for uncommitted changes with 6 specialized reviewer roles (security, bug, quality, contract, testing, history), confidence scoring and false positive filtering
Review Pull Request
Multi-agent PR review system with specialized reviewers, inline comments, and automatic PR description generation using GitHub CLI