Glossary
Eval Coverage Score
A 0–100 score of how well an agent skill has been tested, required for every autonomy promotion.
The Eval Coverage Score measures both breadth and quality of testing for an agent skill across seven weighted dimensions: Contract (15), Skill (20), Safety (20), Tenant (15), Language (10), Simulation (10) and Live production signals (10).
For each dimension, coverage (cases run vs cases required) is multiplied by pass rate; the weighted sum gives the score. Any open critical failure caps the score at 0 for promotion purposes.
Default promotion thresholds are 50 for A1, 70 for A2, 85 for A3 and 92 for A4, each with minimum cases and live samples. A Verified agent template needs at least 85 on its allowed skills across design-partner tenants, or a "Simulator-verified" label until live data exists.
See how One3 would run your business
Tell us about your team and the outcomes you want. We will walk you through the apps, the Digital Employees and the plan for your industry.