Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection when processing PDF documents.
- Ingestion points: The skill extracts content from external PDF files using tools like pdftotext and pdfplumber.
- Boundary markers: Missing delimiters or instructions to treat PDF content as untrusted data.
- Capability inventory: Capability to execute shell commands (qpdf, mutool) and Python code to manipulate files.
- Sanitization: Lacks mention of sanitization or validation of extracted text before processing.
Audit Metadata