baoyu-url-to-markdown

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill launches browser processes such as Chrome, Edge, or Chromium to render web pages before converting them to markdown. Evidence: scripts/cdp.ts uses node:child_process.spawn to start the browser with flags for remote debugging and profile management.
  • [EXTERNAL_DOWNLOADS]: The skill fetches content from user-provided URLs. Evidence: scripts/main.ts accepts a URL argument and navigates to it using CDP.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection from the web content it processes. Ingestion points: Fetches arbitrary content from external URLs (scripts/main.ts). Boundary markers: The output is a markdown document with a YAML frontmatter header but lacks specific delimiters to warn the agent about instructions in the content. Capability inventory: The skill writes to local files in a user-configurable directory. Sanitization: It removes script and style tags before conversion, which prevents execution in the browser but does not filter natural language instructions for the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 03:19 AM
Security Audit — agent-trust-hub — baoyu-url-to-markdown