Prompt Evaluator

A comprehensive framework for evaluating, comparing, and debugging system prompts for AI assistants.

CAPABILITIES:

Single prompt scoring: Evaluate a prompt across 15 dimensions with detailed scores, issues, and specific fix recommendations
Multi-prompt comparison: Compare 2-5 prompts side-by-side with ranking table and improvement roadmap
Feedback analysis: Analyze user feedback, bug reports, or improvement suggestions to determine if issues are prompt-related and provide actionable fixes

TRIGGERS - Use this skill when:

User asks to "evaluate", "score", "rate", or "review" a prompt
User asks to "compare prompts", "which prompt is better", or "A/B test prompts"
User provides feedback/complaints about AI behavior and asks if it's a prompt issue
User asks "what's wrong with this prompt", "how to improve this prompt"
User mentions "system prompt", "inner prompt", "built-in prompt", "assistant prompt"
User asks about prompt quality, prompt optimization, or prompt debugging
Keywords: prompt evaluation, prompt scoring, prompt comparison, prompt analysis, prompt review

prompt-evaluator