canva-prod-checklist

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly calls and ingests responses from the public Canva API (see the canva-prod-verify.sh curl calls to https://api.canva.com/rest/v1/* and the health check fetch to https://api.canva.com/rest/v1/users/me), which can include user-generated design metadata and export URLs that the agent reads and uses to drive subsequent actions (e.g., exports), so untrusted third-party content could influence behavior.