List your company

LLM Agent Security Evaluation on AgentDojo vulnerable-agent profile stress test (six banking pairs)

0Attack Success Count (out of 6)Deny
-0.241.3834.62May 14, 2026

Evaluation results

MethodLinks
DenyDefence=Deny, System=MemLineage2026.050260
Strip+retryDefence=Strip+retry, System=MemLineage2026.0504310
Authority repairDefence=Authority repair, System=MemLineage2026.050444
No defenceDefence=No defence2026.056200
Sig. onlyDefence=Sig. only2026.056200