AgentForge Results Dashboard

Cases per Category

cross-patient-phi2
tool-misuse1
prompt-injection1

Pass / Fail Rate (by category::target_version)

cross-patient-phi::1.0.0-vulnerable0110.00
cross-patient-phi::1.0.1-patched1011.00
tool-misuse::1.0.1-patched1011.00

Resilience Trend (by target_version)

1.0.0-vulnerable0110.00
1.0.1-patched2021.00

Finding Status Counts

open1
in-progress0
resolved1

Cost This Run

0.65

red_team0.42
judge0.08
documentation0.15

Findings Queue (severity order: critical, high, medium, low)

AF-2026-001criticalconfirmed
AF-2026-002mediumneeds-review