scientific-drawing

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute standard LaTeX compilation commands (xelatex) and image conversion tools (pdftoppm, pdftocairo). These are restricted to the figures/ output directory and represent standard functional requirements for generating scientific figures.
  • [EXTERNAL_DOWNLOADS]: No remote scripts or unauthorized external dependencies are downloaded during execution. All operations use local system tools (Python, LaTeX).
  • [DATA_EXFILTRATION]: There are no network requests or attempts to access sensitive system files. All file operations are scoped to the project's ./figures/ directory.
  • [PROMPT_INJECTION]: The skill uses clear, instructional language for task distribution between agents. It does not contain bypass markers, 'ignore instructions' commands, or attempts to override the underlying model's safety guardrails.
  • [COMMAND_EXECUTION]: The skill includes hardcoded local font paths (e.g., /Users/sunyue/Library/Fonts/...) and specific system paths for the author's environment. While these may require adjustment for different users, they do not pose a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 08:21 AM
Security Audit — agent-trust-hub — scientific-drawing