Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Malicious Skill Detection on SKILLVETBENCH labeled set 1.0 (78 malicious + 22 benign skills)
Loading...
0.95
Balance
SKILLVETBENCH
0.4404
0.5727
0.705
0.8373
Jun 14, 2026
Balance
Recall
Precision
Miss Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Balance
Recall
Precision
Miss Rate
SKILLVETBENCH
2026.06
0.95
100
96
0
SkillSieve
2026.06
0.84
85
90
15
SkillProbe
2026.06
0.82
81
88
19
LLM
Prompting strategy=few...
2026.06
0.8
78
88
22
LLM
Prompting strategy=0-shot
2026.06
0.76
74
87
26
CodeBERT
2026.06
0.68
70
95
30
ClawScan
2026.06
0.56
48
100
52
ClawVet
2026.06
0.53
41
100
59
VirusTotal
2026.06
0.46
33
100
67
Feedback
Search any
task
Search any
task