apple-mail
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the macOS 'osascript' utility to execute AppleScript commands for interacting with Mail.app. This allows the agent to programmatically list mail accounts and read message details.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted text from email bodies.
- Ingestion points: 'SKILL.md' (retrieval of message 'content' via AppleScript).
- Boundary markers: The skill does not implement delimiters or 'ignore instructions' warnings for the email content.
- Capability inventory: Capability to execute AppleScript commands via 'osascript'.
- Sanitization: Email content is not sanitized or filtered before being returned to the agent context.
Audit Metadata