write-natural-korean

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses uv run to execute local analysis scripts (article_check.py, lint.py, outline.py, and terms.py). These scripts perform structural and stylistic audits on markdown files to detect issues like duplicate headings, incomplete sections, and repetitive sentence endings.
  • [EXTERNAL_DOWNLOADS]: The analysis scripts depend on the kiwipiepy library, a reputable and widely-used Korean morphological analyzer. This dependency is managed and fetched through the uv package manager from the standard Python Package Index (PyPI).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided article drafts to perform text analysis. It incorporates an adversarial design in its planning phase (references/evidence-integrity.md) by explicitly requiring the agent to distinguish between user-provided facts, verified information, and inferences, which mitigates the risk of the agent adopting false information from processed text as absolute truth.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 11:17 PM
Security Audit — agent-trust-hub — write-natural-korean