Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Security Robustness on DemonAgent
Loading...
0
ASR (w/o)
MCPShield
-1.2
6.9
15
23.1
Feb 15, 2026
ASR (w/o)
ASR (Average)
ASR (Minimum)
ASR (Maximum)
Updated 1mo ago
Evaluation Results
Method
Method
Links
ASR (w/o)
ASR (Average)
ASR (Minimum)
ASR (Maximum)
MCPShield
Backbone=GPT4o-mini
2026.02
0
72
70
80
MCPShield
Backbone=Kimi-K2
2026.02
0
100
100
100
MCPShield
Backbone=Deepseek V3.2
2026.02
0
100
100
100
MCPShield
Backbone=Minimax-M2
2026.02
0
100
100
100
MCPShield
Backbone=Qwen3 235B
2026.02
0
100
100
100
MCPShield
Backbone=Gemini3-Flash
2026.02
30
100
100
100
Feedback
Search any
task
Search any
task
