skills/lgbarn/skills/book-to-research/Gen Agent Trust Hub

book-to-research

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the pdftotext utility to extract text from PDF files located in the ~/Downloads/ directory. This is a standard operation for the skill's documented purpose.
  • [REMOTE_CODE_EXECUTION]: The skill suggests using uv run python with the pymupdf library as a fallback for complex PDF layouts. This involves local execution of Python code for document processing.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from PDF files, creating a surface for indirect prompt injection.
  • Ingestion points: Content extracted from PDFs in ~/Downloads/ (SKILL.md).
  • Boundary markers: None specified in the extraction process.
  • Capability inventory: Local file system writes and Python execution (SKILL.md).
  • Sanitization: No explicit sanitization or filtering of the extracted text is mentioned.
  • [SAFE]: No network exfiltration, hardcoded credentials, or suspicious persistence mechanisms were detected. The skill's behavior aligns with its stated purpose of document research and extraction.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 12:29 AM
Security Audit — agent-trust-hub — book-to-research