latex-paper-en

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs document processing by executing established LaTeX and Typst utilities, such as latexmk, pdflatex, xelatex, and chktex. These operations are handled via Python's subprocess module using argument lists, which prevents shell injection. The skill also includes appropriate safety warnings when the --shell-escape compilation flag is used.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves research metadata from well-known academic services, including CrossRef (api.crossref.org), Semantic Scholar (api.semanticscholar.org), and arXiv (export.arxiv.org). These connections are used solely for citation verification and metadata alignment.
  • [PROMPT_INJECTION]: Instructions within the skill and its reference guides are focused on academic writing tasks and contain no patterns aimed at bypassing safety guardrails or overriding agent behavior.
  • [DATA_EXFILTRATION]: No unauthorized data transmission or suspicious network activity was detected. Network operations are transparently dedicated to obtaining bibliographic data from research repositories.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 02:39 PM