voice-call
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill documents an interface for processing and transmitting voice messages, creating a surface for indirect prompt injection.\n
- Ingestion points: Data enters through the
messageparameter in actions such asinitiate_callandspeak_to_user(SKILL.md).\n - Boundary markers: No delimiters or "ignore" instructions are defined to separate user-provided message content from agent instructions.\n
- Capability inventory: The skill provides the ability to initiate outbound calls and transmit text-to-speech via the
openclawCLI.\n - Sanitization: There is no mention of sanitization or validation of the message content.\n- [COMMAND_EXECUTION]: The skill executes external CLI commands to manage telephony providers.\n
- Evidence: The agent is instructed to use
openclaw voicecallfor placing calls and retrieving status (SKILL.md).
Audit Metadata