quickstart

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

• Third-party content exposure detected (high risk: 0.85). The required workflow ingests outsider-authored free text via the user’s own messages (e.g., provider name, integration ID, and any pasted credentials/API key) into the agent’s LLM context; this is not web-scraped or third-party content, but it is still outsider-authored text because it comes from the operating user’s input rather than the skill’s bundled references.