Cuda Profiling

CUDA profiling skill for NVIDIA GPU performance analysis. Use when profiling kernels with Nsight Systems or Nsight Compute, interpreting roofline models, diagnosing memory-bound vs compute-bound kernels, or annotating code with NVTX ranges. Activates on queries about Nsight, NCU, ncu CLI, GPU roofline, occupancy metrics, or CUDA profiling workflow.

293 stars 36 forksProductivity & Notes

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add mohitmishra786/low-level-dev-skills --skill cuda-profiling
Manual install
git clone --depth 1 https://github.com/mohitmishra786/low-level-dev-skills.git /tmp/low-level-dev-skills
mkdir -p ~/.claude/skills
cp -r /tmp/low-level-dev-skills/skills/gpu/cuda-profiling ~/.claude/skills/cuda-profiling

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# CUDA Profiling

## Purpose

Guide agents through profiling CUDA applications with Nsight Systems (timeline-level) and Nsight Compute (kernel-level metrics), using the NCU CLI for automated metric collection, interpreting roofline models, and diagnosing whether kernels are memory-bound or compute-bound.

## When to Use

- A CUDA kernel is slower than expected and you need bottleneck identification
- Comparing kernel variants (tiling strategies, block sizes)
- Building CI performance regression checks with `ncu` metrics
- Correlating CPU and GPU activity in multi-stream pipelines
- Annotating application phases with NVTX for timeline visibility
- Interpreting occupancy, memory throughput, and SM utilization metrics

## Workflow

### 1. Choose profiling tool

```
What do you need?
├── System-wide timeline (CPU+GPU+CUDA API) → Nsight Systems (nsys)
├── Per-kernel deep metrics (occupancy, memory) → Nsight Compute (ncu)
└── Quick metric from CLI in CI → ncu --metrics ...
```

### 2. Nsight Systems — timeline profiling

```bash
# Profile entire application
nsys profile --trace=cuda,nvtx,osrt --output=report ./my_cuda_app

# Open report
nsys-ui report.nsys-rep

# CLI summary
nsys stats report.nsys-rep
```

What to look for in the timeline:
- Gaps between kernel launches (CPU bottleneck or sync points)
- `cudaDeviceSynchronize` stalls
- Overlap between H2D copies and kernel execution across streams
- CUDA API call overhead
Read the full skill on GitHub

Details

Skill name
cuda-profiling
Path
skills/gpu/cuda-profiling/SKILL.md
License
MIT
Repo updated
14 November 2026

FAQ

What does the Cuda Profiling skill do?

CUDA profiling skill for NVIDIA GPU performance analysis. Use when profiling kernels with Nsight Systems or Nsight Compute, interpreting roofline models, diagnosing memory-bound vs compute-bound kernels, or annotating code with NVTX ranges. Activates on queries about Nsight, NCU, ncu CLI, GPU roofline, occupancy metrics, or CUDA profiling workflow.

How do I install the Cuda Profiling skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add mohitmishra786/low-level-dev-skills --skill cuda-profiling. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Cuda Profiling popular?

It lives in mohitmishra786/low-level-dev-skills, which has 293 GitHub stars and 36 forks.

Related skills

Abi and Calling Conventions

293

mohitmishra786/low-level-dev-skills

ABI and calling conventions skill for cross-language boundaries. Use when explaining System V AMD64, ARM AAPCS, RISC-V psABI, stack frames, variadic calls, or FFI register rules. Activates on queries about calling convention, ABI, System V AMD64, AAPCS, stack frame, variadic function, or FFI registers.

Software Development36

Adc Dac Baremetal

293

mohitmishra786/low-level-dev-skills

Bare-metal ADC and DAC skill. Use when configuring analog sampling, DMA-driven ADC, calibration, or DAC output on MCUs. Activates on queries about ADC bare-metal, sampling time, DMA ADC, or DAC channel setup.

Other36

Af Xdp

293

mohitmishra786/low-level-dev-skills

AF_XDP skill for high-performance XDP sockets. Use when creating AF_XDP sockets, configuring UMEM and XSK rings, XDP_REDIRECT programs, copy vs zero-copy mode, or comparing with DPDK. Activates on queries about AF_XDP, xsk_umem, XDP_REDIRECT, libbpf xsk, or zero-copy XDP.

Software Development36

Dma Baremetal

293

mohitmishra786/low-level-dev-skills

Bare-metal DMA skill for memory-peripheral transfers. Use when configuring DMA channels, circular mode, double buffering, or DMA IRQ completion. Activates on queries about DMA bare-metal, circular buffer, memory-to-peripheral, or DMA stream configuration.

Productivity & Notes36

Cuda

293

mohitmishra786/low-level-dev-skills

CUDA C/C++ skill for NVIDIA GPU kernel programming. Use when writing CUDA kernels, managing thread/block/grid hierarchy, optimizing memory access patterns, using streams and async copies, configuring nvcc flags, or integrating Thrust. Activates on queries about CUDA kernels, nvcc, shared memory, warp divergence, occupancy, or Thrust.

Productivity & Notes36

Email Subject Line Optimizer

171

aiskilloftheweek/claude-ai-skill-of-the-week

Analyzes a subject line against proven best practices and generates 5 optimized alternative versions. Use this skill whenever the user wants to improve, test, or rewrite an email subject line. Trigger on phrases like "improve this subject line", "optimize the subject", "write a better subject line", "suggest subject line alternatives", "this subject line isn't converting", "help me with the subject line", "rewrites for the subject", or when the user pastes an email and asks specifically about the subject line (not the full body). Also trigger when the user uploads or pastes an email and says "

Productivity & Notes35

Browse the most popular Claude skills, all productivity & notes skills, or write your own with the skills generator.