Twilio SMS Automation
Warn
Audited by Snyk on Mar 9, 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 explicitly ingests untrusted, user-generated SMS content via the incoming webhook (see "Handle Incoming SMS" app.post('/sms/incoming') and the two_way_messaging templates that map incoming triggers to actions like lookup_order and opt_out), so inbound messages can be read and acted on by the agent and could therefore carry instructions that influence behavior.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill includes external Twilio webhook/TwiML endpoints (e.g., https://api.example.com/voice/script, https://api.example.com/voice/incoming, https://api.example.com/sms/incoming) that are fetched at runtime by Twilio to obtain TwiML/webhook responses which directly control call/SMS prompts and behavior, making them runtime dependencies.
Audit Metadata