Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt Injection Exposure on AgentDojo
Loading...
48
Source Recall
PI-Hunter
14.096
22.898
31.7
40.502
Jun 10, 2026
Source Recall
Source Diversity
Instruction Recall
Instruction Diversity
Updated 1mo ago
Evaluation Results
Method
Method
Links
Source Recall
Source Diversity
Instruction Recall
Instruction Diversity
PI-Hunter
Defense=None
2026.06
48
42.5
48.6
67.9
PI-Hunter
Defense=Spotlight
2026.06
45.6
38.2
39
47.5
PI-Hunter
Defense=MELON
2026.06
41.7
36.5
40.3
55
Baseline
Defense=None
2026.06
34.8
7.8
27.5
2.5
Baseline
Defense=Spotlight
2026.06
30
0
22
0
PI-Hunter
Defense=PIGuard
2026.06
29.6
31.8
19.4
27.2
Baseline
Defense=MELON
2026.06
28.9
0
42.7
0
Baseline
Defense=PIGuard
2026.06
15.4
0
0
0
Feedback
Search any
task
Search any
task