chrome-devtools-skill

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (SKILL.md) explicitly instructs the agent to automatically navigate to arbitrary public URLs (e.g., "导航到目标 URL" and examples like "抓取 https://news.ycombinator.com" / "打开 https://www.example.com") and then use CDP Runtime.evaluate / evaluate_script to read and extract page content, which is untrusted, user-provided web content that can materially influence subsequent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs "npx -y chrome-devtools-mcp@latest --browser-url=http://127.0.0.1:9222" at runtime, which fetches and executes a remote npm package (chrome-devtools-mcp) that the skill requires to run the MCP server, so it executes remote code during runtime.