critique-to-dex

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

• Third-party content exposure detected (high risk: 0.70). Yes—this skill ingests “findings from any of” including a file path (.md) or pasted text, and then normalizes/extracts fields like issue, impact, evidence, and recommendation into the LLM context to generate Dex tasks, so outsider-authored report text can be included via the runtime “Read it / Parse what is present” path.