style-modeler

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script (normalize_style_frontmatter.py) included in the skill's package. This script is used for normalizing metadata in style files and does not perform network operations or unauthorized file access.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted data from external URLs (specifically WeChat articles).
  • Ingestion points: Data enters via the web-article-extractor tool called on user-supplied URLs in SKILL.md.
  • Boundary markers: None identified; the instructions direct the agent to analyze the extracted content in its entirety.
  • Capability inventory: File write access to .claude/styles/ and execution of the normalize_style_frontmatter.py script.
  • Sanitization: No explicit sanitization or filtering of the extracted article content is performed before the analysis phase.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 08:41 AM
Security Audit — agent-trust-hub — style-modeler