pdfcraft-organize-pdf

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's documentation in references/craft-command-evidence.json references a download URL for a portable binary: https://github.com/storytold/pdfcraft/releases/download/v0.8.0/pdfcraft-0.8.0-windows-x64-portable.zip. This binary is used for executing the document operations described in the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted PDF data while maintaining file-system write capabilities.
  • Ingestion points: Input PDF files are opened and processed via doc_open and doc_info (as described in references/craft-methods.md).
  • Boundary markers: None identified; there are no explicit instructions to the agent to ignore embedded commands within the processed documents.
  • Capability inventory: The skill uses tools like doc_save and page_extract which can write files to the local system.
  • Sanitization: No specific content sanitization or validation logic for the PDF data is mentioned.
  • [COMMAND_EXECUTION]: The skill provides detailed instructions for executing local commands through the pdfcraft-cli tool, including the use of JSON-based scripts to automate multi-step document tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 06:47 PM