playwright

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md and tools-reference explicitly instruct the agent to use browser_navigate and browser_snapshot/browser_evaluate to load and read arbitrary web pages (e.g., "Navigate to a URL and extract content" in SKILL.md and the browser_navigate/browser_snapshot entries in references/tools-reference.md), which exposes the agent to untrusted, user-controlled third-party web content that can influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime installation commands (e.g., "bunx @playwright/mcp@latest" and "npx @playwright/mcp@latest") fetch and execute a remote npm package at runtime to run the Playwright MCP server, which is a required external dependency that executes remote code.