5 agents instrumented across your organization
1 of 4 baselined agents passing
1 learning
across 5 agents
Avg run quality across 5 agents
4 agents requiring action
intent distribution shifted · error rate at 25.0%
intent distribution shifted
intent distribution shifted · score dropped to 62
2 runs recorded but no baseline set
Latest CI activity
Be more helpful to upset customers
failed
Refine deal qualification copy
passed
Add lead_scorer agent (initial cut)
Bump openai-python to 1.51
Tighten support_router escalation rule