skills/iofficeai/aioncore/pdf/Gen Agent Trust Hub

pdf

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process text and data from external PDF files, which could contain malicious instructions intended to manipulate the agent's logic.
  • Ingestion points: Scripts such as extract_form_field_info.py, fill_fillable_fields.py, and merge_pdfs.py read data from local PDF files using the pypdf and pdfplumber libraries.
  • Boundary markers: There are no implementation details for using delimiters or specific system instructions to ignore potential commands embedded within the PDF content.
  • Capability inventory: The skill possesses extensive file system capabilities, including reading and writing PDF, image, and Excel files across multiple scripts.
  • Sanitization: The skill does not appear to perform sanitization or validation of the text extracted from PDF documents before processing it.
  • [COMMAND_EXECUTION]: The skill documentation in SKILL.md and reference.md provides instructions and examples for executing system-level command-line tools such as qpdf, pdftotext, and pdftk. These operations involve shell execution and should be monitored for unexpected command construction.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 08:33 AM
Security Audit — agent-trust-hub — pdf