The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were found. The skill primarily acts as a documentation and configuration interface for an external telephony plugin.\n- [PROMPT_INJECTION]: The skill defines tool actions that ingest untrusted 'message' data for voice output. Ingestion points: 'message' parameter in 'initiate_call', 'continue_call', and 'speak_to_user' (SKILL.md). Boundary markers: Absent. Capability inventory: CLI command execution ('otto voicecall') and telephony network operations. Sanitization: Not specified in the documentation.\n- [COMMAND_EXECUTION]: The skill provides CLI examples using the 'otto' command. The system implementing these tools must ensure the 'message' and 'to' parameters are correctly escaped to prevent shell command injection.

voice-call