Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Malicious Skill Detection on ClawHub Unsafe File Ops 1.0 (n=10)
Loading...
100
Catch Rate
SkillVetBench
16.8
38.4
60
81.6
May 30, 2026
Catch Rate
Correct Alarm
Detection Quality
Miss Rate
Updated 1d ago
Evaluation Results
Method
Method
Links
Catch Rate
Correct Alarm
Detection Quality
Miss Rate
SkillVetBench
2026.05
100
91
95
0
LLM
Evaluation Protocol=fe...
2026.05
80
89
84
20
SkillProbe
2026.05
80
89
84
20
SkillSieve
2026.05
80
89
84
20
LLM
Evaluation Protocol=0-...
2026.05
70
88
78
30
CodeBERT
2026.05
60
100
75
40
ClawScan
2026.05
50
100
67
50
ClawVet
2026.05
40
100
57
60
VirusTotal
2026.05
20
100
33
80
Feedback
Search any
task
Search any
task