Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external, potentially untrusted PDF documents, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill reads and extracts content from PDF files (SKILL.md).
- Boundary markers: There are no markers or instructions defined to prevent the agent from following commands that might be embedded in the PDF content.
- Capability inventory: Includes file system access and shell command execution via CLI tools.
- Sanitization: No sanitization of extracted content is specified.
- [COMMAND_EXECUTION]: The skill describes the use of command-line tools like qpdf and pdftk. While these are standard utilities, the documentation includes examples of passing sensitive data (passwords) as command-line arguments, which is a security risk as it can expose credentials in process lists or shell history.
Audit Metadata