marketing

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

• Third-party content exposure detected (high risk: 0.72). The skill’s workflow explicitly includes harvesting “Competitor…identify via web search if not provided” and then using those search results to ground competitor claims, which can pull outsider-authored free text (web pages/SERP snippets) into the LLM context at runtime via the web-search data path.