imsg

Warn

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the 'imsg' utility from an external third-party Homebrew tap ('steipete/tap/imsg'). This source is not verified as a trusted organization or well-known service.
  • [COMMAND_EXECUTION]: The skill relies on executing system-level commands to query the iMessage database and send messages via the Messages.app automation interface.
  • [PROMPT_INJECTION]: As the skill reads incoming message content, it is susceptible to indirect prompt injection. Malicious instructions sent by external parties via iMessage could influence the agent's behavior. Evidence: Data enters through 'imsg history' (SKILL.md), no boundary markers or sanitization are defined, and the skill has 'imsg send' capabilities.
  • [DATA_EXFILTRATION]: Accesses and processes highly sensitive private communication data, including full chat histories and attachments, which requires Full Disk Access permissions on macOS.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 14, 2026, 12:13 AM