Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt Injection Defense Evaluation on AgentDojo without attack
Loading...
0.49
Total Tokens (M)
tool_filter
0.266
1.778
3.29
4.802
Jun 13, 2025
Total Tokens (M)
Utility
Attack Success Rate (ASR)
Efficiency Score
Updated 23d ago
Evaluation Results
Method
Method
Links
Total Tokens (M)
Utility
Attack Success Rate (ASR)
Efficiency Score
tool_filter
Defense Method=tool_fi...
2025.06
0.49
50.4
7.6
86.6
undefended agent
Defense Method=undefen...
2025.06
0.82
48.3
30.7
21.4
spotlighting_with_delimiting
Defense Method=spotlig...
2025.06
0.88
41
41.8
-0.9
DRIFT
Defense Method=DRIFT
2025.06
2.37
50.9
1.4
20.9
transformers_pi_detector
Defense Method=transfo...
2025.06
2.58
21.2
13
3.2
Progent
Defense Method=Progent
2025.06
2.6
45.6
9.4
13.9
repeat_user_prompt
Defense Method=repeat_...
2025.06
5.43
47.1
15.5
5.8
CaMeL
Defense Method=CaMeL
2025.06
6.09
35.4
0
5.8
Feedback
Search any
task
Search any
task