The Agent Skills Directory

[DATA_EXFILTRATION]: The skill provides the agent with the ability to read private communication history, including message content and attachments, through commands like imsg history and imsg watch. This creates a high-sensitivity data exposure risk if the agent's output is not strictly monitored.
[COMMAND_EXECUTION]: The skill's primary functionality is built upon the execution of the imsg command-line utility to interact with the macOS Messages.app database.
[EXTERNAL_DOWNLOADS]: The instructions direct the user to install an external dependency from a third-party Homebrew tap (steipete/tap/imsg). This introduces a supply-chain risk as the tool is not from an official OS vendor or a recognized trusted organization.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from external incoming messages.
Ingestion points: imsg history and imsg watch commands.
Boundary markers: The instructions do not define delimiters or specific 'ignore instructions' warnings for message content.
Capability inventory: The agent can execute system commands (imsg), read local files (via attachments), and send external messages (imsg send).
Sanitization: No sanitization or validation of the message content is specified before processing.

imessage