nyne

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow calls Nyne API endpoints (e.g., POST /person/search, POST /person/enrichment, POST /company/search) and then ingests the returned JSON results into the agent context; those results are derived from Nyne’s external data sources (outsider-authored content), so the LLM can receive free-text fields from third parties via the API responses.