sales-braze

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly instructs using Connected Content to fetch external APIs at send time (e.g., {% connected_content https://api.example.com/... %} in the Email/Examples and Gotchas sections), so the agent would ingest and act on untrusted third-party API responses as part of its workflow.