imessage
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires installing the imsg utility via a Homebrew tap (steipete/tap/imsg), which is a well-known source for macOS messaging tools.
- [COMMAND_EXECUTION]: Shell commands are used to interact with the imsg CLI for listing chats, retrieving history, and sending messages.
- [PROMPT_INJECTION]: The skill processes untrusted message content, creating an attack surface for indirect prompt injection. 1. Ingestion points: Message history and chat listings are read from the Messages.app database via the imsg CLI (SKILL.md). 2. Boundary markers: Absent; message content is read as raw text. 3. Capability inventory: The skill can send messages via 'imsg send' and perform general shell command execution (SKILL.md). 4. Sanitization: No explicit sanitization or filtering of message content is mentioned, although the skill instructions recommend user confirmation before sending.
Audit Metadata