What is kernel-cuda-specialist?

A specialist for writing raw CUDA C or C++ GPU code in .cu and .cpp files, with pybind11 bindings for use as custom PyTorch extensions. CUDA is NVIDIA’s platform for running code on GPUs.

Use it when you explicitly need .cu or .cpp files compiled through PyTorch’s C++ extension tools.

It provides the right approach when a PyTorch operation needs custom CUDA code instead of a higher-level kernel framework.

How NVIDIA/TensorRT-LLM configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to NVIDIA/TensorRT-LLM. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/NVIDIA/TensorRT-LLM/main/.claude/agents/kernel-cuda-specialist.md

Clone the repo

git clone --depth 1 https://github.com/NVIDIA/TensorRT-LLM

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/nvidia/tensorrt-llm/kernel-cuda-specialist/github.svg)](https://agentmods.dev/agents/nvidia/tensorrt-llm/kernel-cuda-specialist)

Your own site

<a href="https://agentmods.dev/agents/nvidia/tensorrt-llm/kernel-cuda-specialist"><img src="https://agentmods.dev/badge/agents/nvidia/tensorrt-llm/kernel-cuda-specialist/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/nvidia/tensorrt-llm/kernel-cuda-specialist"><img src="https://agentmods.dev/badge/agents/nvidia/tensorrt-llm/kernel-cuda-specialist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 100 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 480 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00100	$0.00480
Opus 5.5	$0.00040	$0.00192
Sonnet 5.5	$0.00020	$0.00096
Haiku 4.5	$0.00010	$0.00048

Measured 17d ago against content hash 342633160dca, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

kernel-cuda-specialist 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 17d ago.

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/agents/kernel-cuda-specialist.md · 46 lines

The source is not reproduced here

A licence we could not identify

The repository carries a LICENSE file, but it is custom or dual enough that GitHub cannot name it and neither can this catalogue. Unknown terms are not permission, so the body is not copied here. Read the licence at the source and decide for yourself.

Read it on GitHub

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.

17d ago Changed · +4 tokens per session 342633160dca
2mo ago First seen · 46 lines · 96 tokens per session scan A 26eb5b152e0f

Subscribe to this mod's changes

kernel-cuda-specialist is an agent published in the GitHub repository NVIDIA/TensorRT-LLM (15,021 stars, last pushed today), with no licence file. It adds 100 tokens to every session and 480 once invoked, about $0.0004 per session on Opus 5.5. 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.

C and C++Languages PyTorch blackwell cuda llm-serving moe

kernel-cuda-specialist next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
kernel-cuda-specialistNVIDIA/TensorRT-LLM	Agent	15k	today	none	A	100	Claude Code
Delphi Expertgithub/awesome-copilot	Agent	41k	yesterday	MIT	A	19	Claude Code
cpp-revieweraffaan-m/ECC	Agent	298k	4d ago	MIT	A	41	Claude Code, Kiro
type-reconstructormarcosd4h/DeepExtractRuntime	Agent	21	1mo ago	MIT	A	63	Claude Code
ciel-systems-guildjxoesneon/Ciel	Agent	1	7d ago	Apache-2.0	A	32	Claude Code
cpp-reviewerloulanyue/awesome-claude-notes	Agent	271	1mo ago	MIT	A	41	Claude Code
cpp-build-resolverFmarzochi/EGC	Agent	68	yesterday	Apache-2.0	A	40	Claude Code

TensorRT-LLM: Agent for Claude Code