market-research-extractor

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

• Third-party content exposure detected (high risk: 0.95). The skill’s runtime workflow explicitly includes “Scrape content based on parameters” from user-specified competitor social/content platforms (e.g., 小红书/B站/抖音/微信/Twitter) and then “Generate research report” using “完整正文,” meaning outsider-authored free text from scraped pages/posts is ingested into the agent’s LLM context.