ClawHub
Benchmarks
TaskDatasetSOTA resultTrendResults
ClawHub
95Overall Detection Rate
9
ClawHub Unsafe File Ops 1.0 (n=10)
100Catch Rate
9
ClawHub Prompt Injection 1.0 (n=19)
100Catch Rate
9
ClawHub Command Injection 1.0 (n=27)
100Catch Rate
9
ClawHub Overall 1.0
95Overall Balance
9
ClawHub
56Top-3 Accuracy
5
ClawHub 0-day
94Win-Rate
2
ClawHub average-day
74.14Win-Rate
1