critique_prompt
Score a prompt on clarity, specificity, and other dimensions, receive per-criterion rationale and concrete suggestions, and get an improved version if the score falls below threshold.
Instructions
LLM-as-judge for a prompt. Scores it 0–10 across 5 default dimensions (clarity, specificity, intent_alignment, format_fitness, length_appropriateness) — or your own custom criteria — and returns per-dimension rationale + concrete suggestions, an overall score, and a verdict (accept / revise / reject). When the score is below revise_threshold (default 7.0), the tool also returns an improvedPrompt you can use as a drop-in replacement. Use it pre-flight (is this prompt good enough for the expensive model?), postmortem (was the prompt the cause of a bad output?), or to A/B-pick the best of N optimization variants. Pass original_prompt when critiquing an optimized version so the judge can verify intent was preserved.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | The candidate prompt to critique. | |
| original_prompt | No | If `prompt` is an optimized version, the user's original ask. Used for the intent_alignment dimension. | |
| category | No | ||
| cwd | No | ||
| file_path | No | ||
| file_language | No | ||
| file_excerpt | No | ||
| user_locale | No | ||
| criteria | No | Override the default 5 criteria. Up to ~8 dimensions; more bloats the judge call. | |
| revise_threshold | No | Overall score below this triggers the rewrite pass. Default 7.0. | |
| skip_rewrite | No | Skip the rewrite pass even when below threshold (faster; just returns scores). |