Product · Version Comparison
Know exactly what changed between v1.42 and v1.43
Baseline against candidate: fixed failures, new regressions and behavior changes, with a clear release verdict.
Headline delta
Reliability score movement with per-metric improved/worsened markers.
Fixed vs new
Which failures the new version resolves, and which ones it introduces.
Behavior changes
Tone, escalation rate and tool usage shifts that no score would surface.
Release verdict
A plain "safe to release" or "blocked" call, backed by the suite.
Test what your agent actually does
Connect an agent, generate scenarios and find the failures before your customers do.