twilio-conversations-classic-api
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements webhook handlers that process incoming message data from external channels, which could potentially contain malicious instructions targeting the logic of the automated agent.
- Ingestion points: External data is ingested via
request.formin the Python Flask snippet andreq.bodyin the Node.js Express snippet withinSKILL.md. - Boundary markers: The example snippets do not implement specific boundary markers or "ignore instructions" delimiters when processing the
EventTypeor message content. - Capability inventory: The skill has the capability to send automated messages using
client.conversations.v1.conversations(...).messages.create(), update conversation states, and manage participants via the Twilio SDK. - Sanitization: The provided code snippets demonstrate basic event type checking but do not include explicit sanitization or filtering of the message content before processing.
Audit Metadata