Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The script
scripts/fill_fillable_fields.pyperforms a runtime monkeypatch of thepypdflibrary. It replaces theDictionaryObject.get_inheritedmethod with a wrapper to resolve a bug in how selection list fields are handled in version 5.7.0 of the library. While functionally justified for the skill's operation, runtime modification of third-party libraries is a notable dynamic execution pattern. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PDF documents and images, creating a surface for indirect prompt injection. 1. Ingestion points: External data enters through
pypdf,pdfplumber,pypdfium2, and OCR tools likepytesseract. 2. Boundary markers: No explicit delimiters or guardrails are defined in the instructions to wrap extracted text content. 3. Capability inventory: The skill performs file system operations (reading and writing PDFs, images, and JSON) and executes helper scripts. 4. Sanitization: Extracted text is not sanitized or filtered for potential malicious instructions before processing. - [COMMAND_EXECUTION]: The instructions and scripts guide the agent to use external command-line tools like
qpdf,pdftotext,pdftk, andpdfimagesfor various document processing tasks.
Audit Metadata