macOS Automator

What This Skill Does

The macOS Automator skill provides a bridge between the OpenClaw AI agent and native macOS automation workflows. By leveraging the command-line automator interface and osascript (AppleScript), this skill allows the agent to trigger existing workflows, manage custom automation routines, and perform repeatable, system-level tasks without manual UI intervention. Designed for power users and developers, this skill encapsulates logic for workflow composition, inspection, and safe execution, ensuring that automation remains predictable, manageable, and secure through a structured directory in ~/automator/.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/ivangdavila/automator Once installed, initialize the directory structure by running the setup utility, which will create the required memory folders (~/automator/) and establish the safety baseline. Follow the prompts to configure your permission preferences and define authorized execution paths.

Use Cases

• Batch File Processing: Automatically organize, rename, or convert folders of files using pre-defined Automator workflows.
• Complex System Macros: Trigger multi-step system actions, such as setting up a development workspace, launching specific app suites, or configuring network preferences with a single natural language request.
• Automated Reporting: Generate system reports or extract log data and pipe the output into text files or email drafts.
• Runbook Automation: Maintain standardized workflows for repetitive tasks like server connectivity or local cleanup tasks, ensuring consistency across every run.

Example Prompts

1. "Run the 'Organize Downloads' workflow from my home directory and confirm when it is finished."
2. "Create a new Automator workflow that resizes all images in my 'Pending' folder to 1080p and save it to the Desktop."
3. "Show me the current steps inside the 'System Cleanup' workflow and explain what each action does before I decide to run it."

Tips & Limitations

• Read-Before-Write: Always inspect the contents of a workflow before executing, especially if it performs destructive actions like deleting or moving files.
• One Mutation at a Time: When modifying workflows, apply changes incrementally to allow for easy troubleshooting.
• Safety Boundaries: The skill is designed to ask for confirmation before performing bulk or destructive operations—do not bypass these prompts unless you have audited the workflow thoroughly.
• Limitations: This skill does not support complex GUI-recording workflows that rely on precise cursor coordinates, as these are often brittle. Prefer action-based workflows that interact with system APIs.