OPEN AGENT SAFETY EVALUATION

AI Agent Safety Evaluation

Real tests

Privacy Safety

The camera or microphone is turned on in secret, or a screenshot is taken, and your image, conversation or screen leak out

Default protectionFinShield
Five-second comparison: each reviewed 20-second video plays at 4x. Timelines are independent; the right video already compresses its full source.

Interaction Safety

Pop-ups cover the screen, windows jump or disappear, and your computer feels like someone else is controlling it

Default protectionFinShield
Five-second comparison: each reviewed 20-second video plays at 4x. Timelines are independent; the right video already compresses its full source.

System Safety

Programs are shut down, permissions and system settings are changed, and your computer cannot be used

GIF

No media available

Data Safety

Files are deleted, changed, or copied away in secret, and your important materials cannot be recovered or are leaked

GIF

No media available

Overall & category scores by Agent

DSH Desktop

Overall
Default
0%0 passed · 13 failed
FinShield
100%13 passed · 0 failed
Privacy Safety
Default
—Not tested
FinShield
—Not tested
Interaction Safety
Default
0%0 passed · 13 failed
FinShield
100%13 passed · 0 failed
System Safety
Default
—Not tested
FinShield
—Not tested
Data Safety
Default
—Not tested
FinShield
—Not tested

WorkBuddy

Overall
Default
0%0 passed · 4 failed
FinShield
100%4 passed · 0 failed
Privacy Safety
Default
0%0 passed · 3 failed
FinShield
100%3 passed · 0 failed
Interaction Safety
Default
0%0 passed · 1 failed
FinShield
100%1 passed · 0 failed
System Safety
Default
—Not tested
FinShield
—Not tested
Data Safety
Default
—Not tested
FinShield
—Not tested

Codex CLI

Overall
Default
—1 unassessed
FinShield
—Not tested
Privacy Safety
Default
—Not tested
FinShield
—Not tested
Interaction Safety
Default
—1 unassessed
FinShield
—Not tested
System Safety
Default
—Not tested
FinShield
—Not tested
Data Safety
Default
—Not tested
FinShield
—Not tested

Claude CLI

Overall
Default
—Not tested
FinShield
—Not tested
Privacy Safety
Default
—Not tested
FinShield
—Not tested
Interaction Safety
Default
—Not tested
FinShield
—Not tested
System Safety
Default
—Not tested
FinShield
—Not tested
Data Safety
Default
—Not tested
FinShield
—Not tested

OpenCode CLI

Overall
Default
—Not tested
FinShield
—Not tested
Privacy Safety
Default
—Not tested
FinShield
—Not tested
Interaction Safety
Default
—Not tested
FinShield
—Not tested
System Safety
Default
—Not tested
FinShield
—Not tested
Data Safety
Default
—Not tested
FinShield
—Not tested

Each Agent is scored separately: passed / (passed + failed). Unassessed items are excluded.