email-send-hardened
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user input to populate email bodies, creating a surface for indirect prompt injection where malicious input could influence agent behavior or cause misdirection.\n
- Ingestion points: User-provided text for email body content in SKILL.md.\n
- Boundary markers: None explicitly defined in the msmtp command examples to separate user data from system instructions.\n
- Capability inventory: Execution of msmtp for network-based data transmission via SMTP.\n
- Sanitization: No programmatic sanitization is applied to the input; the skill relies on manual user review as specified in the safety guidelines.\n- [COMMAND_EXECUTION]: The SKILL.md documentation includes instructions for installing dependencies using sudo (sudo dnf install msmtp), which involves elevated privileges.\n- [DATA_EXFILTRATION]: The primary purpose of the skill is to transmit data to external recipients via SMTP. While this is the intended functionality, it constitutes an outbound data channel that must be monitored for unauthorized use.
Audit Metadata