Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic Security Evaluation on AgentDojo important_instructions attack family v1.1.2
Loading...
2.38
ASR (%)
DRIFT
0.5704
1.0402
1.51
1.9798
Jun 8, 2026
ASR (%)
Exact Count
95% CI (ASR)
Two-sided Fisher p-value
Updated 1mo ago
Evaluation Results
Method
Method
Links
ASR (%)
Exact Count
95% CI (ASR)
Two-sided Fisher p-value
DRIFT
Base Model=gpt-4o-mini...
2026.06
2.38
15
1.34
0.0183
SecureClaw
Base Model=gpt-4o-mini...
2026.06
0.64
4
0.17
-
Feedback
Search any
task
Search any
task