writing-assistant

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

• Third-party content exposure detected (high risk: 0.80). The skill's Integration Map and references/advanced-patterns.md explicitly describe a "Research → Draft" pipeline (using research-assistant and files like research/[topic-slug]-brief.md) where the agent consumes research briefs drawn from public web sources, so it will read and act on untrusted third-party content.