daangn-realty-search

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

• Third-party content exposure detected (high risk: 0.80). The skill directly fetches and parses public, user-generated content from daangn.com (calling the Remix _data search routes and scraping article pages' application/ld+json and as described in SKILL.md and scripts/daangn_realty.py), and that untrusted content is interpreted to produce search/detail results—allowing embedded text from listings to influence the agent's outputs.