baoyu-url-to-markdown

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill launches a local web browser (Chrome, Chromium, or Edge) using child_process.spawn to render target URLs for content extraction. It attempts to find these binaries in standard system paths or respects a user-defined URL_CHROME_PATH environment variable.
  • [PROMPT_INJECTION]: As the skill processes content from arbitrary external URLs, it creates an indirect prompt injection surface. Malicious instructions embedded in a target webpage could potentially influence the agent if it subsequently interprets the generated markdown. The skill mitigates this by using sanitization logic to remove executable tags like <script>, <style>, and <iframe> before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 05:17 AM