seo-growth

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

• Third-party content exposure detected (medium risk: 0.65). The workflow is dominated by first-party “Search Console”/pre-staged data, but the required runtime path explicitly instructs the agent to source free-form “social listening” and repo/release feeds (outsider-authored text) and to use it to “find the exact phrasing people use,” which would then be placed into the agent’s LLM context during the “Mine the freshness gap” step.