Safety
BraveGuard (historical guard decisions)
Published development-checkpoint guard classifications of recorded computer-use-agent trajectories.
536items
3subjects
No license file in the reviewed historical result commits.license
safetydomain
agents_and_tool_usedomain
textmodality
item-level responses released
Saturation status: Yes
Response matrix
Loading response matrix…
Correct (1)Incorrect (0)Unobserved
Scale: 1 = correct · 0 = incorrect