rate_review Prompt Quality Auditor
Tear down uncalibrated prompts against Golden Sets. Use the scorecard to prove calibration value.
description Client's Existing Prompt
Paste the raw system prompt you want to audit.
dataset Golden Set (JSON)
Array of expected_output pairs.
refresh
Simulating edge cases...
Running the prompt against each Golden Set example
error
Audit Failed
— Score
—
analytics Semantic Drift Analysis
—
0
Total Examples
0
Passed
0
Failed
checklist Example Breakdown
Click a row to expand the full reasoning.