markdown-to-html

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs on the installation of markdown tools from official package registries.\n
  • NPM packages: marked, dompurify, sanitize-html, js-xss\n
  • Ruby gems: jekyll, bundler\n
  • Go packages: github.com/gomarkdown/markdown, github.com/gomarkdown/mdtohtml\n- [COMMAND_EXECUTION]: Provides workflows for using command-line tools to process local files.\n
  • Executables: marked, pandoc, hugo, jekyll, go, bundle\n- [PROMPT_INJECTION]: The skill processes untrusted Markdown content which creates an indirect prompt injection surface.\n
  • Ingestion points: Markdown files and strings processed via conversion tools (SKILL.md).\n
  • Boundary markers: Lacks specific delimiters or markers for isolating untrusted data within the agent's processing context.\n
  • Capability inventory: Shell access for file conversion and local web server management.\n
  • Sanitization: Includes explicit recommendations to use sanitization libraries like DOMPurify, sanitize-html, js-xss, and bluemonday.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 11:54 AM
Security Audit — agent-trust-hub — markdown-to-html