data-enrichment

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

• Third-party content exposure detected (medium risk: 0.30). The skill text describes an enrichment workflow where a record-created event triggers an LLM call to “Claude with a website URL” and then parses the LLM response, so the outsider-controlled free text risk is via the scraped website content that is fetched/extracted from arbitrary URLs at runtime.