competitive-brief

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 performs web research (“Company website”, “Recent news”, “Review sites and comparisons”, “News coverage”, etc.), which means it will fetch and ingest outsider-authored free-form text from public web pages and third-party sites into the agent’s LLM context.