deepwork

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 explicitly ingests “confirmed research findings” and “external references” into a local markdown progress file (.slim/deepwork/<slug>.md), and those findings are produced by @librarian from outsider-authored sources (e.g., public web content or other non-user documents) that the LLM then reads/uses as context for later @oracle reviews.