capacity

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

• Third-party content exposure detected (high risk: 0.70). The skill’s REQUIRED workflow includes an “External Research (librarian agent)” track to “find capacity planning best practices” and “research scaling patterns,” which at runtime typically involves fetching public web content or other third-party text that the LLM ingests as free-form prose (outsider source: public web content fetched at runtime via the librarian agent).