termail

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from incoming emails.
  • Ingestion points: External data is read into the agent context through the 'termail list', 'termail read', and 'termail search' commands defined in SKILL.md.
  • Boundary markers: There are no explicit markers or instructions used to separate email content from agent instructions in the command templates.
  • Capability inventory: The agent has access to 'termail compose', 'termail reply', and 'termail forward' capabilities, which could be used to exfiltrate data or send unauthorized communications.
  • Sanitization: No evidence of sanitization or validation of the email body or metadata is present.
  • [EXTERNAL_DOWNLOADS]: The skill relies on an external CLI tool and provides instructions to install it from the author's personal Homebrew tap ('lu-zhengda/tap/termail').
  • [COMMAND_EXECUTION]: The skill executes various system commands through the 'Bash' tool to perform its primary functions, such as syncing and reading email threads.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 03:30 AM