WILD CORPUS · live_scrape
PQS 34 (F) - prompt from docs.anthropic.com
Source: docs.anthropic.com · Scraped 2026-05-04 · Scored 2026-05-04
Score
F34 / 80
cloud_optimize · cloud · pqs-v2.0 · latest (no canonical)
Clarity6 / 10
Specificity5 / 10
Context4 / 10
Constraints7 / 10
Output format3 / 10
Role definition4 / 10
Examples2 / 10
CoT structure3 / 10
Top fixes
- The biggest improvement was adding a structured output format and clear role definition, transforming vague instructions into actionable steps with specific deliverable requirements.
The prompt
Report every issue you find, including ones you are uncertain about or consider low-severity. Do not filter for importance or confidence at this stage - a separate verification step will do that. Your goal here is coverage: it is better to surface a finding that later gets filtered out than to silently drop a real bug. For each finding, include your confidence level and an estimated severity so a downstream filter can rank them.
This prompt was scraped from a public source. The score reflects the input as written, not the quality of any output it produced. The AI input quality problem is the gap between what people type and what the model can act on.