ai302-cli

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

• Third-party content exposure detected (high risk: 0.75). The required workflow includes 302ai search run, whose runtime path returns web-derived free text fields like results[].content/snippet from public web pages fetched by the search provider, which can include outsider-authored content and thus can be used for indirect prompt injection into the agent’s LLM context.