ClarusC64/ai-tool-use-boundary-coherence-risk-v0.1
What this repo is for Detect when AI systems use tools outside allowed boundaries. Focus areas: unauthorized tool execution action without approval escalation failure agent autonomy drift Supports safe deployment of tool-using agents.
019
What this repo is for
Detect when AI systems use tools outside allowed boundaries.
Focus areas:
unauthorized tool execution
action without approval
escalation failure
agent autonomy drift
Supports safe deployment of tool-using agents.
