The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides actions that transmit text messages to external recipients, creating a surface for indirect prompt injection.
Ingestion points: The message parameter in initiate_call and speak_to_user (SKILL.md).
Boundary markers: None present in the tool's interaction logic.
Capability inventory: Initiates outbound calls via Twilio, Telnyx, or Plivo, which could be used for automated social engineering.
Sanitization: Input sanitization is not specified in the skill definition.
[SAFE]: The skill defines configuration requirements for third-party providers (Twilio, Telnyx, Plivo) without hardcoding actual credentials.
[SAFE]: All CLI commands and tool actions are consistent with the skill's stated purpose and do not perform unauthorized system actions.

voice-call