The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill. The functionality matches the stated purpose of providing voice call capabilities.
[COMMAND_EXECUTION]: The skill utilizes the moltbot CLI tool for call operations. This is an expected behavior for a skill designed to extend the Moltbot platform.
[PROMPT_INJECTION]: The skill identifies a data ingestion surface through the message parameter in the voice_call tool actions. This represents a potential surface for indirect prompt injection. Evidence: 1. Ingestion points: message parameter in initiate_call, continue_call, and speak_to_user actions in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: The skill triggers network-enabled telephony plugins via the moltbot CLI. 4. Sanitization: Absent. No specific content filtering or delimiters are specified for the voice messages.

voice-call