agentmods.dev

kernel-cute-writing

A guide for writing GPU kernels with NVIDIA CuTe DSL, the Python interface for CUTLASS 4.x. CuTe DSL provides decorators and layouts for describing GPU kernel code.

In plain words
What is it for?
Writing and implementing GPU kernels with the CuTe DSL and its `cute.jit` and `cute.kernel` constructs.
Why use it?
It keeps CuTe DSL kernel implementation separate from Triton, CUDA C++, and general conceptual questions.

Skill for Claude CodeCodex

15k 3d ago A scan 115 tokens no licence
View the source file NVIDIA/TensorRT-LLM
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/nvidia/tensorrt-llm/kernel-cute-writing
Any agent
npx skills add NVIDIA/TensorRT-LLM --skill kernel-cute-writing
Clone the repo
git clone --depth 1 https://github.com/NVIDIA/TensorRT-LLM

Made for: Claude Code, Codex.

Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,561 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00115 $0.04561
Opus 5 $0.00057 $0.02280
Sonnet 5 $0.00023 $0.00912
Haiku 4.5 $0.00012 $0.00456

Measured 2d ago against content hash 1226645bfd9c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

kernel-cute-writing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/__init__.py, scripts/benchmark_kernel.py, scripts/verify_kernel.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/kernel-cute-writing/SKILL.md · 369 lines

The source is not reproduced here

No licence file

A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.

Read it on GitHub

Files

What ships with it

19 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 369 lines · 115 tokens per session scan A 1226645bfd9c

Subscribe to this mod's changes

kernel-cute-writing is a skill published in the GitHub repository NVIDIA/TensorRT-LLM (14,505 stars, last pushed 3d ago), with no licence file. It adds 115 tokens to every session and 4,561 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

LanguagesPyTorchPythonblackwellcudallm-servingmoe
Related

Other skills, from other repositories

add-or-fix-type-checking

Fixes broken typing checks detected by ty, make typing, or make check-repo. Use when typing errors appear in local runs, CI, or PR logs.

huggingface/transformers · 41 tokens

add-jit-kernel

Step-by-step tutorial for adding a new lightweight JIT CUDA kernel to sglang's jitkernel module.

sgl-project/sglang · 27 tokens

add-sgl-kernel

Step-by-step tutorial for adding a heavyweight AOT CUDA/C++ kernel to sgl-kernel (including tests & benchmarks).

sgl-project/sglang · 31 tokens

add-cuda-kernel

Step-by-step tutorial for adding new CUDA kernels to FlashInfer.

flashinfer-ai/flashinfer · 18 tokens

api-contrib

Contributing to the Kokoro-FastAPI Python API: module layout, endpoint gating pattern, test expectations. Use when adding or changing endpoints, services, or inference code.

remsky/Kokoro-FastAPI · 39 tokens

cuopt-developer

Modify, build, test, debug, and contribute to NVIDIA cuOpt (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions.

NVIDIA/cuopt · 47 tokens