experience-lds-data-requirements-generate

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

• Third-party content exposure detected (medium risk: 0.30). The skill “Generating LDS Data Requirements” consumes an outsider-authored natural-language requirement string supplied by the caller (e.g., PRD snippet / user ask / TODO comment) and parses it in Step 1 (“Parse data requirement”), so the agent ingests free text before selecting specific items.