Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated privacy policy for clearer runtime privacy guarantees and setup requirements. - Improved documentation on system commands and enforced policies (logs-first, allowlisting, rate limiting). - Refined installation instructions for clarity. - No code changes; documentation only.
What This Skill Does
Nexus-Safe monitors local system health and manages service restarts for Docker and PM2 processes. It enforces a logs-first policy and rate limiting before allowing any recovery actions. All operations run entirely on the local machine with no outbound network calls.
Unlike cloud-based monitoring agents, Nexus-Safe keeps all system data local and gates restarts behind mandatory log review and rate limits, reducing the risk of blind restart loops.
When to use it
- Check CPU, RAM, and disk usage on a production server
- Pull recent logs from a crashed Docker container
- Restart a PM2 process after reviewing its error logs
- Monitor load averages during a traffic spike
- Audit which services are allowlisted for auto-recovery
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: /nexus-safe recover my-api
- 1Checks whether my-api is in the NEXUS_SAFE_ALLOWED_PM2 allowlist
- 2Verifies that logs for my-api were retrieved within the last 5 minutes
- 3Checks the sliding window to confirm fewer than 3 restarts occurred in the past hour
- 4Executes the PM2 restart command for my-api
Service my-api restarted successfully; restart count updated in the rate-limit window.
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.