find-flights

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill contains high-risk privacy/backdoor patterns: it explicitly requires loading a local .env from a user path and mandates using a specific personal Chrome profile (on-disk profile, account email, and extension ID) for web scraping and actions, which enables access to stored cookies, sessions, tokens, and other secrets and could be abused to exfiltrate credentials or act on the user's accounts (calendar/bookings) without clear safeguards.

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via runtime-fetched public web content: the skill uses SerpApi (GoogleSearch(...).get_dict()) to ingest Google Flights results (untrusted third-party text) and also uses a live Skyscanner browser session to manually read provider-rendered flight details from the web.