skills.rest
ntk148v
ntk148vCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

system-performance

Diagnose system performance bottlenecks from symptoms to root cause using USE, RED, TSA, and flame graphs.

1|1|Updated Feb 5, 2026
Search Tags:#profiling#latency#performance analysis#flame graphs#USE method#bottleneck diagnosis#Linux perf
One-click install
npx skills add https://github.com/ntk148v/skills --skill system-performance
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: system-performance
Source: https://github.com/ntk148v/skills/tree/main/skills/system-performance
Command: npx skills add https://github.com/ntk148v/skills --skill system-performance

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you systematically diagnose slow systems and performance regressions by turning vague “it’s slow” complaints into measurable bottleneck hypotheses.

Core Features & Use Cases

  • Structured bottleneck discovery (USE Method): Evaluate utilization, saturation, and errors across resources such as CPU, memory, disk, and network.
  • Service-level latency diagnosis (RED Method): Analyze request rate, error rate, and latency duration percentiles for services and microservices.
  • Thread and CPU hot-spot visibility (TSA + Flame graphs): Break down time spent across OS thread states and visualize CPU stack hotspots to find what code paths consume time.
  • Workflow support: Provides problem-statement framing, workload characterization, drill-down analysis, and elimination-based root-cause testing, plus anti-methodologies to avoid.

Quick Start

Ask your AI to apply the USE method to identify CPU, memory, disk, or network bottlenecks and then use flame-graph guidance to pinpoint the hot code paths causing the slowdown.

Frequently Asked Questions about system-performance

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I diagnose system performance bottlenecks using the USE method?▼

To diagnose system performance bottlenecks with the USE method, systematically evaluate the utilization, saturation, and errors of resources like CPU, memory, disk, and network to narrow down root causes.

What is the best way to find CPU hotspots using flame graphs and Linux perf?▼

Finding CPU hotspots with flame graphs and Linux perf involves breaking down time spent across OS thread states and visualizing CPU stack traces to pinpoint exactly which code paths consume the most time.

How do I troubleshoot service latency and microservices performance issues?▼

Troubleshoot service latency and microservices performance by applying the RED method to analyze request rate, error rate, and latency duration percentiles, providing actionable troubleshooting workflows.

Can I use this systematic drill-down approach for disk I/O and network saturation?▼

Yes, you can use this systematic drill-down approach for disk I/O and network saturation, applying stepwise elimination procedures and metric selection to test bottleneck hypotheses across computing environments.

What should I avoid when doing performance analysis to prevent wrong conclusions?▼

When doing performance analysis, avoid anti-methodologies that lack systematic metric selection or workload characterization, which can lead to wrong conclusions and prevent actionable root-cause testing.

Related Skills

Explore More Software Engineering Skills→
sahiixxsahiixx

system-info

Collect CPU, memory, disk, network, process, and environment metrics via Node.js scripts.

Community
Basic
JylhisJylhis
1

m10-performance

Identify performance bottlenecks in Rust projects and recommend optimizations.

Community
Intermediate
avifeneshavifenesh
951

perf-profiler

Profile CPU and memory hotspots to generate flame graphs and JFR/perf evidence.

Community
Intermediate
dralgorhythmdralgorhythm
111

observability

Implement centralized logging, metrics, and distributed tracing with OpenTelemetry.

Community
Advanced