browser-automation

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to ask for user credentials and to "browser_type" the password field (and warns only not to store them), which requires the model to receive and inject secret values verbatim into actions/outputs, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md workflows and the agent system_prompt explicitly direct use of browser_navigate, browser_read_page, web_search, and web_fetch to visit and interpret arbitrary public websites (e.g., "Product Search & Purchase", "Price Comparison" and Phase 2/3 steps), so it ingests untrusted third‑party web content that can influence subsequent clicks, form fills, and decisions.