Agent skill · data analytics · aiskillstore
gasp-diagnostics
System diagnostics using GASP (General AI Specialized Process monitor). Use when user asks about Linux system performance, requests system checks, mentions GASP, asks to diagnose hosts, or says things like "check my system" or "what's wrong with [hostname]". Can actively fetch GASP metrics from hosts via HTTP or interpret provided JSON output.
Why this skill is useful
Provides specific commands and workflows for fetching and interpreting GASP metrics that the AI wouldn't reliably generate on its own.
What it needs
About 4k tokens when loaded. Last updated 2026-08-07. 405 stars on the source repository.
What this skill does
GASP Diagnostics Enables comprehensive Linux system diagnostics using GASP's AI-optimized monitoring output. Actively fetches metrics from hosts and provides intelligent analysis with context-aware interpretation. Fetching GASP Metrics When user mentions a host or requests a system check: 1. Fetch the metrics endpoint 2. Hostname formats supported mDNS/local: accelerated.local, hyperion.local DNS names: proxmox1, dev-server, workstation IP addresses: 192.168.1.100 3. Default port: 8080 (unless user specifies otherwise) 4. Error handling Host unreachable: Inform user, suggest checking if GASP is running Port closed/refused: Try suggesting systemctl status gasp on the host JSON parse error: GASP may not be installed or wrong endpoint Timeout: Network issue or host down 5. Multi-host queries: If user mentions multiple hosts, fetch each in sequence and compare Quick Diagnosis Workflow For any system check request: 1. Fetch metrics from specified host(s) 2. Check summary first: Look at summary.health and summary.concerns[] 3. Identify issues using metric correlations below 4. Report findings with severity and specific recommendations Trigger Examples These user messages should trigger this skill and active fetching: "Check hyperion for me" "What's going on with accelerated.local?" "Is proxmox1 having issues?" "Compare hyperion and proxmox1" "Why is my system slow?" (fetch localhost) "Diagnose 192.168.1.50" "Check all my proxmox nodes" Metric Interpretation Health Summary summary.health: Quick assessment "healthy": No action needed "degraded": Issues present but not critical "critical": Immediate attention required summary.concerns[]: Pre-analyzed issues to investigate first summary.recentchanges[]: Context for current state CPU Analysis Load ratio = loadavg1m / cores: < 0.7: Normal usage 0.7-1.0: Busy but healthy 1.0-2.0: Saturated (may cause slowness) \> 2.0: Severe overload Key indicators: trend: "increasing" is concerning even if current load is acceptable baselineloa …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aiskillstore/gasp-diagnostics