content-factory

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

• Third-party content exposure detected (high risk: 0.90). This skill's SKILL.md and data files show it automatically fetches public, user-generated content from open platforms (see scripts/aggregator/fetch_all.py and the data/hotpool/*.json listing Weibo, Zhihu, YouTube, GitHub, Douyin, etc.) and then feeds that content into downstream LLM-driven steps (scripts/topic_scorer.py, scripts/content_generator.py) that influence topic selection, generation, and publishing—meeting the criteria for exposure to untrusted third-party input that could enable indirect prompt injection.