academic-latex-pipeline

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Markdown content to generate LaTeX documents.\n
  • Ingestion points: Reads and processes SourceDocument.md and project sections.\n
  • Boundary markers: No explicit markers are used to separate untrusted content from the LaTeX instructions.\n
  • Capability inventory: Employs pdflatex, bibtex, and fc-cache for compilation and font setup, alongside file system access in the build script.\n
  • Sanitization: The build script performs formatting cleanup but does not sanitize against LaTeX command injection or shell escape sequences.\n- [COMMAND_EXECUTION]: The workflow involves executing multiple shell commands for document processing.\n
  • Evidence: Commands such as pdflatex, bibtex, and fc-cache are invoked in the build process described in SKILL.md.\n
  • Note: These are standard operations for the skill's primary function.\n- [EXTERNAL_DOWNLOADS]: The documentation points to a well-known service for downloading required fonts.\n
  • Evidence: References github.com/googlei18n/noto-cjk/releases for CJK font support.\n
  • Note: This reference is considered safe as it targets an official repository from a well-known provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 11:42 AM