Keamanan
No visual example yet
Explore the skill
agent-safety
ericrisco
Use when bounding an LLM agent that already runs — scoping its task domain, gating tools to least privilege, defending against prompt injection in untrusted web/email/RA…
58GitHub