ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified system Safety 4/5

system-watchdog

System resource monitoring. Detects actionable anomalies (memory pressure, runaway processes, disk pressure) and reports only when something needs attention. Optimized for few, high-signal alerts. Works on both Linux and macOS.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/brennerspear/system-watchdog
Or

What This Skill Does

The system-watchdog skill is a robust, lightweight monitoring agent designed for macOS environments. It operates by scanning active system processes and resource consumption metrics to identify inefficiencies, runaway tasks, or potentially problematic background operations. Unlike broad monitoring tools that provide overwhelming telemetry, system-watchdog is purpose-built to filter the "noise" and highlight only what truly matters: processes that are leaking memory, consuming excessive CPU, or running stale for unreasonable durations. By outputting structured JSON, the skill is perfectly optimized for autonomous AI agents, allowing them to make data-driven decisions about whether to notify the user or perform self-corrective actions like terminating a zombie process.

Installation

To install this skill, use the OpenClaw CLI package manager. Run the following command in your terminal:

clawhub install openclaw/skills/skills/brennerspear/system-watchdog

Once installed, you can integrate it into your automated workflows. For recurring monitoring, it is highly recommended to add the provided script to your crontab. Refer to the openclaw.md file within the repository for specific instructions on setting up a persistent background watch schedule that runs check.sh every hour.

Use Cases

  • Autonomous Cleanup: An AI agent monitors for 'stale' dev processes (like forgotten local node servers or hanging python scripts) and automatically kills them to free up RAM.
  • Performance Alerts: Receive concise notifications when a specific process (e.g., an LLM inference server) spikes above defined memory thresholds.
  • System Health Reporting: Generate automated morning reports summarizing disk usage and system load to ensure your machine is ready for a heavy day of development.

Example Prompts

  1. "Check the current system health. If any processes are using excessive memory, list them and suggest which ones I should terminate."
  2. "Run the system-watchdog and provide a summary of the current resource load and any suspicious processes found."
  3. "Is my current development environment bloated? Use system-watchdog to see if there are any stale processes older than two days."

Tips & Limitations

  • Platform Compatibility: Note that this skill is currently optimized for macOS. If you are operating on Linux, the script requires modification of the underlying shell commands (replacing sysctl/vm_stat with /proc filesystem calls).
  • Grace Period: If you encounter a JSON parse error during the initial execution, do not panic; it is often a race condition with the process scanner. The agent should be configured to retry once before flagging an error.
  • Thresholds: You can inspect check.sh to adjust the threshold values. If you are working with large datasets, you might want to increase the RAM limit from the default 4096MB to avoid false positives.

Metadata

Stars4190
Views1
Updated2026-04-18
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-brennerspear-system-watchdog": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#system#monitoring#automation#macos#performance
Safety Score: 4/5

Flags: file-read, code-execution