render-markdown

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes @tanstack/markdown, a package from a recognized developer. All implementation examples follow standard usage patterns without suspicious behaviors.\n- [SAFE]: The library's security-first design is documented, specifically highlighting that unsafe URL protocols are stripped and raw HTML is ignored unless explicitly enabled.\n- [INDIRECT_PROMPT_INJECTION]: The skill establishes a pathway for processing untrusted Markdown content.\n
  • Ingestion points: Strings provided to parseMarkdown and renderHtml (SKILL.md, references/ast-and-options.md).\n
  • Boundary markers: The library requires allowHtml: true to recognize HTML nodes, defaulting to a safer state.\n
  • Capability inventory: The tool is designed for string transformation; it does not perform network requests or file modifications.\n
  • Sanitization: Sanitization of link and image protocols is performed automatically during parsing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 08:33 AM
Security Audit — agent-trust-hub — render-markdown