trtllm-moe-develop

A guide for reviewing and changing mixture-of-experts (MoE) code in TensorRT-LLM. MoE models use different groups of model components, called experts, for different inputs.

In plain words
What is it for?
Planning, reviewing, refactoring, or implementing changes to TensorRT-LLM PyTorch MoE modules and backends.
Why use it?
It helps keep MoE changes aligned with the surrounding architecture and easier to maintain and test.

Skill for Claude CodeCodex

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/trtllm-moe-develop
Any agent
npx skills add NVIDIA/TensorRT-LLM --skill trtllm-moe-develop
Clone the repo
git clone --depth 1 https://github.com/NVIDIA/TensorRT-LLM

Made for: Claude Code, Codex.

Per session 149 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,678 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 $0.00149 $0.07678
Opus 5 $0.00075 $0.03839
Sonnet 5 $0.00030 $0.01536
Haiku 4.5 $0.00015 $0.00768

Measured 3d ago against content hash 12d32361eaec, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

trtllm-moe-develop 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 3d 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/skills/trtllm-moe-develop/SKILL.md · 680 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

1 file 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. 3d ago First seen · 680 lines · 149 tokens per session scan A 12d32361eaec

Subscribe to this mod's changes

trtllm-moe-develop is a skill published in the GitHub repository NVIDIA/TensorRT-LLM (14,505 stars, last pushed 3d ago), with no licence file. It adds 149 tokens to every session and 7,678 once invoked, about $0.0007 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.