Build Your Evaluation Pack
Create a complete evaluation pack for your agent: define test categories, set thresholds, configure regression checks, and run the full suite to generate a quality report.
Evaluation Pack Builder
Build your evaluation pack step by step. Define categories, set pass/fail thresholds, add regression checks, then run the evaluation suite.
Define Test Categories
Add categories for your evaluation pack. Each category groups related test cases. Use the presets or create your own.
Set Pass/Fail Thresholds
For each category, set the minimum pass rate (percentage of tests that must pass for the category to be considered acceptable).
Configure Regression Checks
Select which categories should include regression monitoring. Regression checks flag when a previously passing test starts failing between versions.
Run Evaluation Pack
Your evaluation pack is ready. Click Run to execute all test cases and generate the quality report.