article-writing

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

• Third-party content exposure detected (high risk: 1.00). The skill's "Voice Capture Workflow" explicitly instructs the agent to collect and ingest published articles, newsletters and X/LinkedIn posts (public, user-generated content) to extract tone, meaning the agent will read and act on untrusted third-party content such as social media and public articles.