baoyu-danger-gemini-web

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The code deliberately reads browser authentication cookies (from profiles and via a Chrome CDP session), persists them to local cookie files, and uploads local files to external endpoints — actions that access and could exfiltrate sensitive credentials and user data and therefore present a high credential-theft / data-exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text via user-supplied CLI prompt (including --promptfiles) and sends it to Gemini in scripts/main.ts → GeminiClient.generate_content(prompt, ...).