perf-review

Profile code to identify runtime performance bottlenecks and ranked hotspots.

3|Updated Mar 1, 2010

Search Tags:#performance #profiling #baseline #measurement #code-analysis #hotspots

Ready to run

npx skills add https://github.com/harleypig/dotfiles --skill perf-review-harleypig

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: perf-review
Source: https://github.com/harleypig/dotfiles/tree/main/config/claude/skills/perf-review
Command: npx skills add https://github.com/harleypig/dotfiles --skill perf-review-harleypig

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Identify runtime performance bottlenecks by profiling code and surfacing hotspots for targeted optimization.

Core Features & Use Cases

Baseline-based profiling to reveal hottest paths, memory usage, and I/O hotspots.
Measure-first workflow that separates confirmed findings from hypotheses.
Scalable to codebases of varying size and complexity, from single services to multi-service architectures.

Quick Start

Run a baseline profiler on the target codebase and report the top hotspots ranked by measured impact.

Frequently Asked Questions about perf-review

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

FAQPage Schema

How do I identify runtime performance bottlenecks in my code?▼

You identify runtime performance bottlenecks by profiling code against a baseline to surface hotspots. This measure-first workflow delegates profiling to a subagent and outputs ranked, actionable findings for targeted optimization.

What is the best way to profile code and find slow paths during development?▼

The best way to profile code and find slow paths is using a measure-first workflow that separates confirmed findings from hypotheses. It establishes a baseline, reveals the hottest paths, and detects quantifiable slowdowns across services.

Can I use this profiling approach across multi-service architectures?▼

Yes, this profiling approach is scalable and applies across multi-service architectures. It effectively detects memory usage and I/O hotspots across libraries and endpoints, handling codebases of varying size and complexity.

How does baseline-based profiling reveal memory usage and I/O hotspots?▼

Baseline-based profiling reveals memory usage and I/O hotspots by comparing current runtime metrics against an established baseline. This method isolates the hottest paths and quantifiable slowdowns, separating confirmed findings from hypotheses.

Do I need to establish a baseline before profiling code for hotspots?▼

Yes, you must establish a baseline before profiling code for hotspots. The workflow requires a baseline measurement to accurately detect slow paths and quantify slowdowns, ensuring findings are ranked by measured impact.

What are the limitations of a measure-first approach to code analysis?▼

A limitation of this measure-first code analysis is that it separates confirmed findings from hypotheses, meaning speculative optimizations are deprioritized until runtime profiling data validates the quantifiable slowdowns.