lov-illustrate

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses curl and wget to download images and official assets from the web for inclusion in documents.
  • [COMMAND_EXECUTION]: Executes local Python utilities audit_html_fonts.py and profile_store.py to perform typography audits and manage user settings.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided Markdown files, which may contain embedded instructions that influence the illustration planning process.
  • Ingestion points: Content from the target Markdown file path.
  • Boundary markers: Analyzes full document text without explicit isolation markers.
  • Capability inventory: Network retrieval, file system writes, headless browser interaction, and local configuration management.
  • Sanitization: Performs semantic analysis without explicit sanitization of document text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 10:34 AM
Security Audit — agent-trust-hub — lov-illustrate