unified-docs
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified in the skill instructions, templates, or scripts.
- [COMMAND_EXECUTION]: The skill employs local Python scripts (scripts/cascade_targets.py and scripts/check_frontmatter.py) to automate document auditing. These scripts include an extensive list of excluded paths to ensure the agent does not process sensitive tool-specific or configuration directories.
- [DATA_EXFILTRATION]: There are no network calls or patterns indicating that sensitive data is being sent to external servers. All operations are strictly local.
- [PROMPT_INJECTION]: The skill enforces a 'First User-Gated' policy and a 'Discovery Gate' for all creation tasks. This architecture ensures the agent remains aligned with user intent and provides a defense against potential injection from processed documentation by requiring the agent to validate its understanding with the user before acting.
Audit Metadata