authoring-user-guide

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text via its research step: it uses deep-research or falls back to web search, and consumes external search-result text for facts (even though screened for prompt injection), which is then fed into the agent’s LLM context.