review-paper

Pass

Audited by Gen Agent Trust Hub on Oct 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a structured workflow for generating academic manuscripts using local templates (references/macro_skeleton.md).
  • [SAFE]: The bundled verification scripts (tests/check_skeleton.py and tests/test_macro_skeleton.sh) perform static structural analysis of templates without performing network requests or accessing sensitive system files.
  • [SAFE]: Instructions enforce rigorous reference verification, mandating that citations resolve to a project-specific bibliography file (_src/refs.bib) to prevent citation fabrication and hallucination.
  • [SAFE]: The skill explicitly restricts its file-writing operations to a designated manuscript directory, adhering to the principle of least privilege for file system access.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 4, 2026, 01:27 PM