skills/teamwiseflow/xiaobei/email-ops/Gen Agent Trust Hub

email-ops

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a shell wrapper (email-ops.sh) to execute a Python script (send_email.py). This is a standard and safe way to expose skill functionality to the agent environment.
  • [DATA_EXFILTRATION]: While the skill can send data via email (SMTP), it is the primary purpose of the skill. The implementation uses secure connections (STARTTLS/SSL) and acts only upon explicit user instruction. No unauthorized data exfiltration patterns were detected.
  • [CREDENTIALS_UNSAFE]: The skill correctly manages sensitive credentials by requiring SMTP_USER and SMTP_PASSWORD to be provided via environment variables, adhering to secure development practices.
  • [REMOTE_CODE_EXECUTION]: No remote code execution or untrusted package installation patterns were found. The skill relies on Python's built-in standard libraries (smtplib, ssl).
  • [PROMPT_INJECTION]: The skill uses structured prompts to guide the LLM in drafting emails. These prompts do not contain instructions to bypass safety filters or ignore system rules.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 06:38 AM
Security Audit — agent-trust-hub — email-ops