Prompt Quality Scorer — Score Any Prompt 0-100

Paste any AI prompt. Get an instant 0-100 quality score across 6 dimensions, plus specific suggestions to improve. Calibrated to the 2026 prompt-engineering scaffold.

The 6 dimensions we score

  • Role definition (15 pts): Is there a specific role with credentials? "Senior security engineer with 10 years" beats "expert."
  • Principles (15 pts): Are quality bars + constraints encoded? Numbered principles outperform vague guidance.
  • Input structure (20 pts): Is input wrapped in tags or clearly labeled? <input> tags beat inline mixing.
  • Output format (20 pts): Is the output structure specified? "Output JSON: {...}" or markdown headings.
  • Examples (15 pts): Few-shot examples present for tone/format you can't describe?
  • Constraints + negative space (15 pts): "Do NOT..." patterns and explicit refusals.

What good prompts look like

Browse 588+ Promptolis Originals for fully-structured examples. Each follows the 4-block scaffold: role + principles + input + output-format. Score them in this tool — most land 85-100.

Related tools

Token Counter →

Count tokens before sending.

Cheat Sheet PDF →

10 patterns + 20 do's/don'ts.

Browse Originals →

588+ structured prompts.

Get new tools + Originals every Friday

2-3 hand-crafted Originals + occasional new tools. No spam, unsubscribe in 1 click.

Or subscribe directly on Beehiiv →