seo-ai-search-share-of-voice

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text is ingested via the runtime MCP data calls DATA_getAiPromptsByBrand / DATA_getAiPromptsByTarget, which return sampled ChatGPT prompt text and cited sources authored by third parties (i.e., non-operating-user content) that the skill then saves into 03-prompts-{domain}.md and uses for topic clustering.