file-naming
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted file content using OCR and text extraction tools to determine naming conventions, which creates an attack surface where instructions embedded in files could attempt to influence the agent's behavior.
- Ingestion points: File content analysis step in
SKILL.mdand episode lookup inreferences/tv-episodes.md. - Boundary markers: No specific delimiters or instructions are provided to the agent to disregard content that looks like commands.
- Capability inventory: File system write access (renaming).
- Sanitization: Not explicitly defined for the extraction process.
- Mitigation: The risk is significantly reduced by the required user confirmation step and the plan validation phase that prevents accidental collisions or overwrites.
Audit Metadata