parallel-data-enrichment

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

• Third-party content exposure detected (high risk: 0.70). This skill explicitly performs "web-sourced" enrichment and its required workflow (SKILL.md) tells the agent to preview rows of the output CSV produced by that enrichment, so the agent will read untrusted, public third‑party content as part of runtime behavior which could enable indirect prompt injection.