playwright-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill contains examples that embed credentials verbatim (e.g., cookie-set with "abc123", fill "password123") and shows capturing a cookie/token to stdout (TOKEN=$(playwright-cli --raw cookie-get session_id)), which requires the agent to read and potentially emit secret values directly.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md describes opening user-provided/selected URLs (outsider-controlled web pages) at runtime and then feeding the resulting screenshot-derived inputs (screenshot_paths -> visual-mode runs “as if the user uploaded screenshots manually”) into the agent LLM context, which can include free text extracted/encoded from those pages.