Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to use external command-line utilities including qpdf, poppler-utils (pdftotext, pdfimages), and ImageMagick to perform advanced document manipulation tasks.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests untrusted content from external PDF files (documented in scripts/extract_form_structure.py and scripts/extract_form_field_info.py). There are no explicit boundary markers or sanitization routines for the extracted data, which the agent uses to perform form-filling operations. Capabilities include file system access and execution of command-line tools.\n- [SAFE]: The provided Python scripts rely on established libraries like pypdf, pdfplumber, reportlab, and Pillow, and do not contain hardcoded credentials or unauthorized network exfiltration logic.\n- [SAFE]: A localized monkeypatch in scripts/fill_fillable_fields.py modifies the pypdf library at runtime to address specific limitations in handling form field attributes, which is a legitimate functional improvement.
Audit Metadata