perf-profile

Analyzes codebases for performance bottlenecks and generates prioritized optimization reports against budgets.

Updated Apr 13, 2026
One-click install
npx skills add https://github.com/onemanking/the-operator --skill perf-profile-onemanking
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perf-profile
Source: https://github.com/onemanking/the-operator/tree/main/.github/skills/perf-profile
Command: npx skills add https://github.com/onemanking/the-operator --skill perf-profile-onemanking

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Performance issues are often discovered too late, and gut-feel optimization wastes effort on the wrong code. This Skill provides a structured static-analysis workflow that identifies CPU, memory, rendering, and I/O hotspots, measures them against defined performance budgets, and produces a prioritized optimization report. ## Core Features & Use Cases - Scoped Profiling: Profile a single named system or run a comprehensive full-codebase analysis via the argument hint. - Budget Comparison: Reads target FPS, memory, load time, and draw call budgets from design docs and flags each metric as OK, WARNING, or OVER. - Prioritized Report: Generates a Markdown report with hotspots (file:line, impact, fix effort), ranked recommendations, quick wins, and areas needing runtime profiling. - Use Case: A game developer notices frame drops before launch. Invoke the skill with 'full' to get a table of hotspots in per-frame update loops, allocation patterns, and draw call estimates, plus the top three fixes ranked by expected gain. ## Quick Start Ask the AI to run a performance profile on the rendering system and report the top hotspots against the 60fps frame budget.

Frequently Asked Questions about perf-profile

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

FAQPage Schema
How do I profile my codebase for performance bottlenecks?

Invoke the skill with a system name to scope the analysis, or use 'full' for a comprehensive profile. It examines per-frame update functions, allocation patterns, physics queries, and rendering costs, then outputs a prioritized hotspot report.

How to check if my game meets a 60fps frame budget?

The skill reads performance budgets from design docs or CLAUDE.md, such as a 16.67ms frame budget for 60fps. It estimates current cost per metric and marks each as OK, WARNING, or OVER in the generated report.

Can static analysis replace runtime performance profiling?

No. Static analysis identifies candidate hotspots like nested loops and per-frame allocations, but the skill explicitly flags areas requiring runtime profiling to confirm actual impact. Measure on target hardware before optimizing.

What performance issues does the analysis detect?

It targets CPU issues like expensive per-frame functions and string operations in hot paths, memory issues like leaks and missing object pooling, rendering issues like overdraw and missing LODs, and I/O issues like synchronous asset loading.

When should I not optimize based on this report?

Avoid optimizing without runtime measurement, since static estimates can misrank true bottlenecks. The report distinguishes CPU-bound, GPU-bound, and I/O-bound candidates so you confirm the actual constraint first.