paper-deck

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content such as arXiv papers, DOI links, and PDF files to generate slide summaries and visual prompts. While this presents an ingestion surface for indirect prompt injection, the skill utilizes a multi-stage workflow (analysis, brief, outline) that involves explicit confirmation steps, effectively limiting the risk of accidental instruction following from source documents.
  • [COMMAND_EXECUTION]: The skill employs a local Python script (scripts/merge_deck.py) to compile generated images into PPTX and PDF formats. This script relies on reputable libraries such as python-pptx and Pillow and operates solely on local file paths without performing network operations or incorporating unsanitized user input into system commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:49 AM