file-manager
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes file names and directory structures, which are external data sources that could theoretically contain hidden instructions.
- Ingestion points: Local file system metadata including file names and directory paths (SKILL.md).
- Boundary markers: The skill explicitly requires showing previews and obtaining user confirmation before executing moves or deletions.
- Capability inventory: File and folder operations including creation, moving, renaming, and recursive deletion.
- Sanitization: The skill includes a 'Security Rules' section that specifically forbids operations on system directories (/System, /usr, /bin) and files belonging to other users.
- [SAFE]: The skill follows security best practices for file management by implementing scope restrictions to prevent modification of operating system files and requiring human-in-the-loop confirmation for high-impact actions.
Audit Metadata