nature-writing

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill provides structured guidance and templates for scientific writing, including abstracts, introductions, and submission materials. All analyzed content is transparent and strictly aligned with the stated purpose of assisting in manuscript preparation.
  • [NO_CODE]: The skill consists entirely of Markdown instructions and LaTeX templates. It does not include executable scripts, binary files, or automated installation procedures. References to Python libraries in documentation are provided for diagnostic purposes only and are not executed by the skill itself.
  • [PROMPT_INJECTION]: No evidence of prompt injection, jailbreak attempts, or instructions designed to bypass agent safety filters was found. The instructions emphasize fidelity to author-provided evidence and explicitly forbid the invention of results.
  • [DATA_EXFILTRATION]: No network operations, hardcoded credentials, or instructions to access sensitive local files (such as environment variables or SSH keys) were detected. All external references target reputable documentation or related writing tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 08:48 AM