nsys-ai
GPU profile analysis, bottleneck diagnosis, and run comparison
All Skills in This Repository (1)
Pure Emerald Level IndicatorsFrequently Asked Questions
FAQPage SchemaHow to install nsys-ai?▼
Run `npx skills add GindaChen/nsys-ai --all -g -y` in your terminal to install all skills in this suite globally.
How to find GPU training bottlenecks?▼
Point the skill at your .sqlite or .nsys-rep profile and it runs automated triage to identify idle gaps, NCCL stalls, or kernel hotspots, then reports one root cause with a fix.
Can nsys-ai compare two profiling runs?▼
Yes. The diff mode compares before and after profiles, reports regressions and improvements, and gives a verdict on whether your change actually helped.
Does nsys-ai work with Claude Code and Cursor?▼
Yes. The skills follow the universal SKILL.md standard and run in Claude Code, Cursor, and other compatible coding agents.
Do I need a GPU or CUDA installed to analyze profiles?▼
No. Analyzing an existing .sqlite profile only requires Python 3.10 or later; a GPU and Nsight are only needed to capture new profiles.
Related Repositories in Software Engineering
View All in Software Engineering→
openclawopenclaw
Run a personal AI assistant across your devices and chat apps
obrasuperpowers
Gives coding agents a disciplined workflow from idea to merged code
facebookreact
AI agent skills for building, testing, and porting React core