skills/rheinmir/setup/check-approve/Gen Agent Trust Hub

check-approve

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided parameters like request codes and container names to construct shell command templates. * Ingestion points: REQ-CODE, be-container, and user parameters in SKILL.md. * Boundary markers: Absent. * Capability inventory: File writing to the scratchpad/ directory. * Sanitization: No validation of input strings is performed, which could lead to command injection in the generated script if the user runs it without review.
  • [DYNAMIC_EXECUTION]: The skill generates a .sh shell script file from predefined templates. This is categorized as low risk because the agent is specifically instructed not to execute the script, requiring manual user intervention.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 02:19 AM
Security Audit — agent-trust-hub — check-approve