why

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

• Third-party content exposure detected (high risk: 0.75). At runtime the skill queries multiple MCP-backed systems (issue/ticket tracker, long-form docs, real-time team chat, observability, error tracking, analytics) and ingests their returned free-text content into the synthesizer/LLM context; those sources can include outsider-authored text (e.g., external users/customers, third-party commenters, or non-operating-user engineers) via authenticated APIs.