send-sms
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates the transmission of user-provided content to external SMS gateways, which is a potential vector for indirect prompt injection if instructions are embedded in the message. * Ingestion points: The 'message' and 'to' parameters in SKILL.md accept arbitrary strings. * Boundary markers: No specific delimiters or instructions to treat the message body as untrusted content are provided to the agent. * Capability inventory: The skill performs network POST requests to external APIs and user-defined custom URLs. * Sanitization: While phone number format and message length are validated, no semantic content sanitization or escaping is described.
- [DATA_EXFILTRATION]: The skill is designed to transmit data (phone numbers and message bodies) to external endpoints. It utilizes well-known service providers like Twilio and Vonage for its core functionality. It also supports 'Custom Provider Setup' which allows the agent to send data to arbitrary, non-whitelisted HTTP endpoints defined in the provider_config.url parameter.
Audit Metadata