ts-agent-sdk
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill includes generic HTTP clients that fetch and return data from arbitrary external URLs (templates/api/base.ts request + templates/api/webhook.ts WebhookClient.get/post) and also directly calls public third‑party APIs (templates/api/public/holidays.ts calling the Nager.Date API), so the agent can ingest untrusted, user-controlled web content which it will read/interpret as part of its workflows.
Audit Metadata