perf-profile

Profile systems against performance budgets and generate ranked optimization reports.

Updated Apr 4, 2026
One-click install
npx skills add https://github.com/WindXRan/y3td --skill perf-profile-windxran
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perf-profile
Source: https://github.com/WindXRan/y3td/tree/main/.claude/skills/perf-profile
Command: npx skills add https://github.com/WindXRan/y3td --skill perf-profile-windxran

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you identify performance bottlenecks by measuring them against explicit frame, memory, load-time, draw-call, and related budgets, then produces a prioritized optimization plan with clear next steps.

Core Features & Use Cases

  • Profiling scope control: Focuses profiling on a specific system name or runs a full cross-system analysis when you pass full.
  • Budget-driven reporting: Loads performance targets (FPS, memory, load time, draw calls, and networking limits) from existing docs and generates a structured report that marks OK/WARNING/OVER status.
  • Actionable optimization output: Produces hotspots plus ranked recommendations (with expected gain, risk, and fix approach), followed by a decision phase to choose implement/defer/escalate options based on fix effort.

Quick Start

Run perf-profile with the relevant system name (or use full) to generate a performance report with prioritized hotspots and recommendations against your budgets.

Frequently Asked Questions about perf-profile

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

FAQPage Schema
How do I identify performance bottlenecks in my codebase against a specific memory or frame-time budget?

To identify performance bottlenecks, profile targeted systems and compare results against declared performance budgets. This generates a structured report marking OK, WARNING, or OVER status for memory, frame-time, and resource constraints.

What is the best way to profile CPU and rendering hotspots across a full codebase?

The best way to profile CPU and rendering hotspots across a full codebase is to run a full cross-system analysis. This applies frame-time and resource constraints to produce ranked optimization recommendations with estimated impact and fix effort.

Can I focus performance profiling on a single subsystem instead of the entire project?

Yes, you can focus performance profiling on a single subsystem by providing the specific system name. This narrows the analysis to produce targeted hotspots and optimization recommendations for that subsystem.

How do I prioritize optimization plans based on fix effort and expected gain?

You prioritize optimization plans by using a structured report that ranks recommendations based on expected gain, risk, and fix approach. A decision phase then helps choose implement, defer, or escalate options based on fix effort.

Does performance profiling work with existing FPS and draw-call target documents?

Yes, performance profiling works with existing documents by loading performance targets like FPS, load time, and draw-call limits. It uses these budgets to generate a structured report marking status against your declared limits.

What should I do when performance profiling marks a subsystem as OVER my declared budget?

When profiling marks a subsystem as OVER your declared budget, follow the next-step decision flow to choose implement, defer, or escalate based on the ranked optimization recommendations and estimated fix effort.