system_monitor

What This Skill Does

The system_monitor skill is a diagnostic tool designed for OpenClaw agents to gain real-time visibility into the hardware health of the host machine. By executing a local shell script, the agent can retrieve precise metrics regarding CPU load, physical memory consumption, and GPU utilization. This capability is essential for agents tasked with maintaining server stability, monitoring intensive computational processes, or simply providing status updates to administrators. By abstracting the complexity of command-line monitoring tools, this skill provides a clean, structured output that the AI agent can parse and report back to the user in plain English.

Installation

To integrate this monitoring capability into your OpenClaw environment, execute the following installation command in your terminal:

clawhub install openclaw/skills/skills/bobdevibecoder/bobagent-system-monitor

Ensure that you have the necessary execution permissions for the resulting ./monitor.sh script, as the skill requires read access to system statistics files located in the root directory.

Use Cases

This skill is primarily used by system administrators and developers who need to keep a close eye on resource-heavy tasks. Use cases include:

• Proactively checking for resource bottlenecks before launching large training jobs.
• Debugging server slowdowns or unresponsive processes.
• Providing automated status dashboards for long-running background tasks.
• Verifying that GPU hardware is correctly allocated and utilized during machine learning model inference.

Example Prompts

1. "Check the current system health. Are we hitting our CPU limits?"
2. "How much RAM is currently being used, and is the GPU under heavy load?"
3. "Run a system check and tell me if the hardware is running too hot or at maximum capacity."

Tips & Limitations

The system_monitor skill relies on the presence of the monitor.sh script at the expected file path. If your environment does not support shell script execution, this skill may fail. Additionally, the skill provides a snapshot in time; it does not currently support long-term trend logging or historical data analysis. For best results, ensure your user profile has the appropriate permissions to query system-level metrics, as restricted user environments may prevent the script from accessing sensitive hardware information. Regularly update the skill via clawhub to receive compatibility patches for newer OS kernels.