daangn-cars-search

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and parses public, user-generated listings from daangn.com (see "Data surfaces" in SKILL.md: the region resolver and /kr/cars/?_data routes, and scripts/daangn_cars.py's fetch_json calls and carPost detail fetch), so untrusted third‑party content is ingested and used to drive search/detail behavior and could carry indirect prompt-injection content.