imsg
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs an external binary
imsgvia a third-party Homebrew tap (steipete/tap/imsg). While the source is a known developer, it is not within the pre-defined list of trusted organizations. - [COMMAND_EXECUTION]: The skill relies on executing the
imsgCLI tool to perform actions like listing chats, reading history, and sending messages. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted text data from external sources (SMS and iMessages).
- Ingestion points: The
imsg historyandimsg watchcommands (referenced inSKILL.md) allow the agent to read message content from external senders. - Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from obeying commands embedded within the chat history.
- Capability inventory: The agent has the capability to send outgoing messages and files using
imsg sendbased on interpreted data. - Sanitization: There is no evidence of sanitization or filtering applied to the message content before it is processed by the agent.
Audit Metadata