imsg

What This Skill Does

The imsg skill for OpenClaw provides a robust command-line interface to interact with the native macOS Messages.app. By leveraging this tool, users can perform essential messaging tasks directly from their terminal or via an AI agent. It supports listing active chat threads, retrieving full message histories including media attachments, watching for incoming messages in real-time, and sending both text and file-based multimedia messages. This skill bridges the gap between your automated workflows and your personal or professional messaging streams, allowing for seamless integration of SMS and iMessage capabilities within an intelligent agent framework.

Installation

To integrate this skill into your environment, ensure you have the OpenClaw agent installed. Run the following command in your terminal to fetch the package from the official registry: clawhub install openclaw/skills/skills/minsu7893/imsg-1-0-0

Prerequisites include having your Apple ID signed into Messages.app, granting your terminal emulator 'Full Disk Access' via System Settings (to allow read access to the local SMS database), and providing 'Automation' permissions to allow the tool to trigger send events within the system.

Use Cases

• Automated Alerts: Send status updates or error logs from server monitoring scripts directly to your phone.
• Context Retrieval: Fetch the last 20 messages from a specific conversation to provide an AI agent with context before responding to a user query.
• File Sharing: Automate the distribution of reports, screenshots, or generated media files to specific contacts.
• Live Monitoring: Monitor specific high-priority chat threads for incoming requests or information, triggering downstream agent tasks based on keywords.

Example Prompts

1. "Check my latest messages from 'Mom' and summarize the most recent request she made."
2. "Send the file at /Users/me/reports/quarterly_results.pdf to +14155551212 with the message 'Here is the requested report.'"
3. "Watch for any incoming messages in the work chat and alert me if someone mentions the word 'urgent'."

Tips & Limitations

• Safety First: Always confirm the recipient identifier and the message body before the AI agent triggers the execution of a 'send' command to prevent accidental messages.
• Permission Management: If the skill fails to list chats, verify that your terminal has Full Disk Access; macOS is strict about permissions for the Messages database.
• Service Selection: Use the --service flag to specify iMessage, SMS, or 'auto' to ensure the message routes correctly based on the recipient's capability.
• Rate Limiting: Avoid high-frequency sending to prevent potential flagging by Apple's internal messaging throttling mechanisms.