mobile-forensics

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes examples that directly embed plaintext credentials (e.g., parser.decrypt(password="backup_password")) and instructs requesting/providing backup passwords/tokens, requiring the agent to handle and potentially output secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.80). The skill directly ingests and analyzes untrusted, user-generated third‑party content — e.g., messaging/social data (MessagingAnalyzer.parse_whatsapp, parse_telegram, parse_messenger, parse_instagram), cloud data (CloudAnalyzer.analyze_google, analyze_icloud, get_google_location_history, extract_cloud_tokens), and browser histories (parser.get_chrome_history, parser.get_safari_history) — which the agent is expected to read and interpret.