openclaw-ops

What This Skill Does

The openclaw-ops skill is the essential diagnostic and recovery toolkit for maintaining an OpenClaw Gateway environment on a local machine. It functions as a surgical instrument for rescue agents tasked with resolving downtime or performing preventive maintenance. Rather than managing application-level logic, this skill operates at the infrastructure layer, handling service lifecycle management, log analysis, and connectivity verification on both Linux (systemd) and macOS (launchd) platforms.

Installation

To integrate this operational toolkit into your agent environment, execute the following command in your terminal:

clawhub install openclaw/skills/skills/dinstein/openclaw-ops

Ensure that your environment has the necessary permissions to interface with service managers like systemctl or launchctl to allow the skill to inspect and modify gateway states.

Use Cases

This skill is engineered for two primary operational scenarios:

1. Emergency Rescue: When your gateway service is unresponsive, crashing, or returning errors, use this skill to diagnose root causes—such as port conflicts, memory exhaustion (OOM), or corrupted configuration—and perform corrective actions to restore the gateway to an operational state.

2. Routine Health Check: During standard maintenance, utilize this skill to monitor resource consumption, inspect recent logs for hidden warning signs, and verify network availability on your configured ports. It helps ensure the gateway maintains optimal performance without human intervention.

Example Prompts

• "The gateway is currently down and won't restart; please run a full diagnostic, check the logs for errors, and fix the service."
• "Perform a health check on the OpenClaw gateway to ensure it is listening on the correct port and report any recent crash logs."
• "I suspect a port conflict is preventing the gateway from starting. Identify what is running on the configured port and resolve the issue."

Tips & Limitations

• Principle of Operation: Always follow the 'Diagnose → Judge → Act → Verify' workflow. Do not attempt to fix a state you haven't fully diagnosed.
• Platform Awareness: The skill is platform-aware, but ensure your environment variables are set correctly so the skill accurately identifies between systemd and launchd.
• Scope: Do not use this for application-level issues like prompt tuning or skill management; this is strictly for gateway service integrity.
• Port Detection: Always rely on the dynamic port detection feature rather than assuming a default value, as custom configurations are common in production environments.