gpc-app-recovery
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected in the skill analysis.
- [COMMAND_EXECUTION]: The skill uses the
gpccommand-line tool to list, create, and manage app recovery actions, which aligns with its documented purpose. - [PROMPT_INJECTION]: The skill defines a surface for processing external data via JSON files. Evidence Chain: 1. Ingestion point: File system read of
recovery.jsonandtargeting.jsoninSKILL.md. 2. Boundary markers: Agent instructions require displaying the draft and obtaining user confirmation before deployment. 3. Capability inventory: Execution ofgpccommands through a subprocess. 4. Sanitization: Not explicitly detailed, but the human-in-the-loop requirement provides a critical security checkpoint.
Audit Metadata