superseller-higgsfield-discipline

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

• Third-party content exposure detected (high risk: 0.90). This skill's Research stage explicitly instructs the agent to "Pull content_competitors rows for tenant + Apify competitor scrape if recent (<7d)" (Stage 1 — Research), meaning it ingests public/scraped competitor social/web content which the agent must read and synthesize to drive planning and generation, so untrusted third-party content can influence subsequent actions.