skills.rest
xchang1121
xchang1121COMMUNITY
Explore all skills by this author →
GitHub

performance-summary

Aggregate performance metrics from verify_<op_name>/result.json files across operators.

6|1|Updated Apr 19, 2026
Search Tags:#performance#metrics#reporting#profiling#benchmark#verify#compare
One-click install
npx skills add https://github.com/xchang1121/AutoResearch-CC-hook --skill performance-summary
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: performance-summary
Source: https://github.com/xchang1121/AutoResearch-CC-hook/tree/main/skills/performance-summary
Command: npx skills add https://github.com/xchang1121/AutoResearch-CC-hook --skill performance-summary

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill aggregates and compares performance results across multiple operators to help you quickly identify bottlenecks, rank implementations, and decide optimization priorities.

Core Features & Use Cases

  • Collect performance metrics from verify_<op_name>/result.json and optional profiling.json.
  • Generate comparison tables and a concise performance report.
  • Analyze results to highlight outliers, compute speedups, and suggest targeted optimizations.

Quick Start

Run the aggregate tool to summarize results across targeted operators and produce a unified report.

Frequently Asked Questions about performance-summary

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

FAQPage Schema
How do I consolidate performance metrics across multiple operators?▼

To consolidate performance metrics across multiple operators, collect verify_<op_name>/result.json and profiling.json data to generate comparison tables and a concise performance report highlighting bottlenecks and speedups.

What is the best way to compare kernel performance results for benchmarking?▼

The best way to compare kernel performance results for benchmarking is to aggregate verify_<op_name>/result.json data to rank implementations, identify the best performers, and produce a unified performance summary with optional visuals.

Do I need profiling.json files to generate a performance summary?▼

You do not need profiling.json files to generate a basic performance summary, but accessible verify_<op_name>/result.json files are required. Profiling.json is only necessary for deeper performance analysis.

How do I identify optimization priorities from operator benchmarking results?▼

To identify optimization priorities from operator benchmarking results, analyze the aggregated metrics to highlight outliers and compute speedups, which suggests targeted optimizations for the bottleneck implementations.

Can I generate visual performance reports from multiple operator result.json files?▼

You can generate visual performance reports from multiple operator result.json files by running the aggregate tool to summarize results across targeted operators, producing a structured performance summary with optional visuals.

Related Skills

Explore More Data & Analytics Skills→
mindspore-aimindspore-ai
258

performance-summary

Summarize and compare operator performance results from result.json files.

Official
Intermediate
FuncSmileFuncSmile

V3 Performance Optimization

Optimize Claude v3 performance with benchmarking and Flash Attention acceleration.

Community
Advanced
AscendAscend
31

triton-operator-precision-eval

Benchmark Triton operator precision against PyTorch references and generate MERE/MARE reports.

Official
Advanced
jdforsythejdforsythe

wolfe-perf

Detect and report performance inefficiencies using static analysis, benchmarking, and profiling.

Community
Advanced