skills.rest
sqjian
sqjianCOMMUNITY
Explore all skills by this author →
GitHub

generate-profile

Launch an SGLang server, validate accuracy, and generate a Chrome/Perfetto trace.

Updated May 14, 2026
Search Tags:#performance#profiling#trace#server#gpu#profile#sglang
One-click install
npx skills add https://github.com/sqjian/sglang --skill generate-profile-sqjian
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: generate-profile
Source: https://github.com/sqjian/sglang/tree/main/.claude/skills/generate-profile
Command: npx skills add https://github.com/sqjian/sglang --skill generate-profile-sqjian

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables end-to-end performance profiling of an SGLang server by launching the server, validating its accuracy, and generating a Chrome/Perfetto-compatible trace, returning the profile path for analysis.

Core Features & Use Cases

  • End-to-end profiling: Launches the server, runs a quick accuracy check, and captures a profiling trace.
  • Trace outputs: Produces a timestamped trace directory containing a gzipped trace file and server configuration.
  • Use Case: Ideal for developers optimizing model-serving pipelines, GPU utilization, and latency profiling.

Quick Start

Launch the SGLang server, run the profiling workflow, and return the path to the generated trace.

Frequently Asked Questions about generate-profile

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

FAQPage Schema
How do I capture an end-to-end profiling trace for an SGLang server?▼

To capture an SGLang server profiling trace, you need an automated workflow that launches the server, validates accuracy, and generates a Perfetto-compatible trace file for latency and GPU optimization analysis.

What do I need to generate a Chrome trace for GPU performance profiling?▼

Generating a Chrome trace for GPU performance profiling requires a working SGLang installation and at least one CUDA-capable GPU to execute the server, run validation checks, and output the trace directory.

Can I measure startup and validation latency during SGLang server profiling?▼

Yes, SGLang server profiling measures startup, validation, and trace collection latency across GPU-enabled environments, outputting a timestamped directory containing the gzipped trace and server configuration.

What is the best way to profile model-serving pipelines for GPU utilization?▼

The best way to profile model-serving pipelines for GPU utilization is to run an end-to-end profiling workflow that launches the server, executes a quick accuracy check, and captures a Perfetto-compatible trace.

What outputs do I get from an automated SGLang server profiling workflow?▼

An SGLang server profiling workflow outputs a timestamped trace directory containing a gzipped Chrome/Perfetto trace file and a server_args.json configuration file, returning the profile path for analysis.

Related Skills

Explore More Software Engineering Skills→
Fridge003Fridge003
9

generate-profile

Generate a Chrome-compatible profiling trace for an SGLang server run.

Community
Advanced
drunkcodingdrunkcoding
4

generate-profile

Profile an SGLang server end-to-end and generate Chrome-compatible trace files.

Community
Advanced
FutureMLS-LabFutureMLS-Lab
556

generate-profile

Generate end-to-end performance profiling traces for SGLang server runs.

Official
Intermediate
dyyounggdyyoungg

generate-profile

Profile SGLang server runs and capture Chrome-compatible traces for performance analysis.

Community
Advanced