voice-call
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates voice calls where the agent may process external speech or call data, creating a surface for injection.
- Ingestion points: The
continue_callandget_statusactions inSKILL.mdingest transcriptions or metadata from the external call environment. - Boundary markers: No explicit delimiters or boundary markers are defined to isolate incoming call data from the agent's internal instructions.
- Capability inventory: The skill can initiate, continue, and monitor voice calls, performing network operations via external telephony providers.
- Sanitization: No explicit sanitization or filtering of call content is mentioned in the skill definition.
- [DATA_EXFILTRATION]: The skill initiates network operations to external telephony services including Twilio, Telnyx, and Plivo. These interactions are fundamental to the skill's intended purpose of managing voice calls and rely on configuration settings (account SIDs and API tokens) for these well-known communication platforms.
Audit Metadata