perf-profiler

Profiles CPU and memory hotspots in applications and outputs flame graphs and traces.

1.9k|545|Updated Oct 22, 2025
One-click install
npx skills add https://github.com/ComposioHQ/awesome-claude-plugins --skill perf-profiler-composiohq
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perf-profiler
Source: https://github.com/ComposioHQ/awesome-claude-plugins/tree/main/perf/skills/profile
Command: npx skills add https://github.com/ComposioHQ/awesome-claude-plugins --skill perf-profiler-composiohq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps identify performance bottlenecks by profiling CPU and memory hotspots in applications, providing evidence in the form of flame graphs and other artifacts.

Core Features & Use Cases

  • Profiling CPU and memory hotspots to locate bottlenecks in hot paths.
  • Evidence Flame graphs, JFR/perf traces, and structured outputs for verification.
  • Use Case Imagine a long-running service with sporadic latency; use this Skill to capture hotspots and generate traceable performance evidence.

Quick Start

Run the profiler on a target command using the provided tool hint, for example: perf-profiler <tool> <command>

Frequently Asked Questions about perf-profiler

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

FAQPage Schema
How do I profile CPU hotspots to identify performance bottlenecks?

To profile CPU hotspots, you run the profiler on a target command using the provided tool hint. It captures execution data to locate performance bottlenecks in hot paths and generates traceable evidence like flame graphs.

What is a flame graph and when do I need it for performance profiling?

A flame graph is a visual artifact representing CPU and memory usage during execution. You need it during performance profiling to quickly identify hot paths and sporadic latency in long-running services or batch jobs.

Can I use this profiler for long-running services with sporadic latency?

Yes, you can profile long-running services with sporadic latency. The profiler captures hotspots in hot paths and generates structured outputs and flame graphs for verification.

What's the best way to generate traceable evidence for memory hotspots?

The best way to generate traceable evidence for memory hotspots is to run the profiler on your target command. It produces JFR or perf traces and flame graphs to verify performance bottlenecks.

Do I need standard profiling workflows to capture batch job hotspots?

Yes, capturing batch job hotspots requires integration with standard profiling workflows. You run the profiler on the target command to generate flame graphs and other structured artifacts.