file-organization
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local filesystem, such as filenames and shebang lines, which could be used to influence the agent's classification logic. \n
- Ingestion points: Scanning target directories and collecting metadata (names, hashes, interpreter paths) as defined in
SKILL.md(Workflow Step 1). \n - Boundary markers: Employs a mandatory dry-run preview and user confirmation checkpoint before any modifications are made in
SKILL.md(Workflow Step 5). \n - Capability inventory: Extensive filesystem access including directory creation, file moving, renaming, and archiving. \n
- Sanitization: Filenames are sanitized via normalization to kebab-case or snake_case and special character removal in
SKILL.md(Workflow Step 4). \n- [COMMAND_EXECUTION]: The skill calls system-level utilities for specific environment integrations. \n - Evidence: The use of
git mvto preserve history in version-controlled directories is specified in theBest Practicessection ofSKILL.md.
Audit Metadata