telnyx-messaging-java
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external messaging services which could be used to influence the agent's behavior.
- Ingestion points: Inbound message content is processed via the
message.receivedwebhook event, specifically thedata.payload.textfield as described inSKILL.mdandreferences/api-details.md. - Boundary markers: The skill explicitly documents how to perform signature verification using
client.webhooks().unwrap()to ensure data authenticity. However, it does not suggest specific delimiters or instructions for the agent to ignore potentially malicious content within the message body itself. - Capability inventory: The skill possesses significant capabilities, including sending outbound SMS/MMS/WhatsApp messages (
client.messages().send()), scheduling messages, and managing messaging profile settings. - Sanitization: While signature verification is correctly implemented to validate the source, the skill lacks explicit guidance on sanitizing the actual string content of inbound messages before it is passed to the LLM context.
Audit Metadata