imsg
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the
imsgbinary via a third-party Homebrew tap (steipete/tap/imsg) which is not an official vendor resource. - [DATA_EXFILTRATION]: The skill provides tools to read highly sensitive private data (iMessage and SMS history). It requires "Full Disk Access" permissions on macOS to read the Messages database, exposing personally identifiable information.
- [COMMAND_EXECUTION]: Utilizes the
imsgCLI to perform system interactions, including database queries and message transmission. - [INDIRECT_PROMPT_INJECTION]: The skill reads external message content which could contain malicious instructions designed to influence the agent's behavior. * Ingestion points: Message history and live watch streams accessed via
imsg historyandimsg watch(SKILL.md). * Boundary markers: Absent. No specific delimiters or instructions to ignore commands within the message text are provided. * Capability inventory: Sending messages and attachments viaimsg send(SKILL.md). * Sanitization: Absent. External message content is processed directly into the agent's context.
Audit Metadata