Security audit
Relational Safety for Agents
Security checks for vulnerabilities and agentic risk
Overview
This is a self-contained safety-curriculum skill whose optional external actions are disclosed and require human approval.
Before installing, understand that this skill changes the agent's conversational operating style around consent, boundaries, and safety. Optional GitHub posting or installing the linked follow-up skill should only happen if you explicitly approve it.
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.