file-organization
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use various system utilities for file analysis and management, including
pdftotextfor content extraction,lsoffor checking active file handles, andstat/dufor metadata verification. It also involves standard shell-level operations for moving and deleting files during the consolidation process. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by scanning filenames and extracting text from documents to identify consolidation candidates, creating a potential injection surface. * Ingestion points: Filenames retrieved from the filesystem and text content extracted from PDFs and other documents via
pdftotextas specified inSKILL.md. * Boundary markers: The skill does not define explicit delimiters or 'ignore instructions' warnings for when document content is interpolated into the agent's context. * Capability inventory: The skill possesses read access to the filesystem, the ability to move or rename files, and the ability to delete byte-identical duplicates. * Sanitization: The procedure mandates manual review of false positives and borderline files before any destructive or relocation operations occur, which serves as a human-in-the-loop safeguard against adversarial content triggers.
Audit Metadata