skills/isomoes/skills/pdf/Gen Agent Trust Hub

pdf

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PDF documents which presents a potential vulnerability surface for instructions embedded in data.\n
  • Ingestion points: Untrusted data is ingested via PDF files in scripts/extract_form_field_info.py, scripts/fill_fillable_fields.py, and scripts/fill_pdf_form_with_annotations.py.\n
  • Boundary markers: No explicit prompt boundary markers are used; the skill relies on instructional workflow guidance in forms.md.\n
  • Capability inventory: The skill possesses capabilities for file system operations, execution of PDF utility shell commands, and runtime library modification.\n
  • Sanitization: Employs geometric validation for bounding boxes in scripts/check_bounding_boxes.py and field value validation in scripts/fill_fillable_fields.py.\n- [DYNAMIC_EXECUTION]: The script scripts/fill_fillable_fields.py performs runtime monkeypatching on the pypdf library to fix a known issue with handling selection list fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:16 AM
Security Audit — agent-trust-hub — pdf