twilio
Warn
Audited by Snyk on Nov 21, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The required
twilio api core messages listworkflow (andmessages fetch) reads recent messages, which include inbound SMS bodies that any outsider can send by texting a Twilio number, without the agent selecting a specific item first.
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (medium risk: 0.30). Decisive operation:
twilio phone-numbers buy --country-code US --type local(line 44) purchases a phone number, which is a direct purchase. Sending SMS (line 51) and placing calls (line 60) are billable usage, which can also be read as indirect operations. Under the recall-first override, I treat the purchase as an in-scope direct purchase rather than a pure service cost. Control: the documentedtwilio login(lines 24-26) authenticates the account session that the CLI uses to execute the purchase. That is a service-enforced login gate, so the direct route has qualifying access control and is medium rather than high. The skill does not document any per-purchase approval. Reading logs and listing resources are not financial operations.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (low risk: 0.10). The skill exposes operational changes, such as buying phone numbers, updating a number's SMS webhook, and installing or removing CLI plugins, but no explicit irreversible operational loss is established, so it stays at low.
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata