peer-review

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute local Python scripts (scripts/generate_schematic.py and skills/scientific-slides/scripts/pdf_to_images.py) to perform complex tasks like publication-quality diagram generation and PDF-to-image conversion.
  • [PROMPT_INJECTION]: The skill carries a risk of indirect prompt injection because it is designed to analyze untrusted external manuscripts and presentation files.
  • Ingestion points: Untrusted data enters the agent's context through the reading of manuscript files and the conversion of presentation PDFs into images for inspection.
  • Boundary markers: The skill lacks clear delimiters or 'ignore' instructions to distinguish the document content from the agent's system instructions.
  • Capability inventory: The agent is equipped with several powerful tools, including Bash, Read, Write, and Edit, which could be exploited if an injection is successful.
  • Sanitization: No sanitization or content validation procedures are defined for the information extracted from the processed documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:26 PM