neuron-framework-autoport-vllm-neuron

neuron-framework-autoport-vllm-neuron is a skill for Claude Code from aws-neuron/neuron-agentic-development. It costs 40 tokens per session (7,847 once invoked), scanned B, original, Apache-2.0.

A workflow for adapting a Hugging Face machine-learning model architecture to the vLLM-Neuron backend on AWS Trainium2. vLLM is software for serving language models, and a backend is the part that connects it to specific hardware.

In plain words
What is it for?
Use it to port a model into vLLM-Neuron, place the model code and example in the expected project folders, and optionally perform a dry run without compiling or running inference.
Why use it?
It covers the research, code changes, and checks needed when a model is not yet supported by that Trainium2 backend.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: positional $N argument.

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/aws-neuron/neuron-agentic-development/neuron-framework-autoport-vllm-neuron
Any agent
npx skills add aws-neuron/neuron-agentic-development --skill neuron-framework-autoport-vllm-neuron
Clone the repo
git clone --depth 1 https://github.com/aws-neuron/neuron-agentic-development

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.

agentmods badge for neuron-framework-autoport-vllm-neuron

README.md
[![agentmods](https://agentmods.dev/badge/skills/aws-neuron/neuron-agentic-development/neuron-framework-autoport-vllm-neuron.svg)](https://agentmods.dev/skills/aws-neuron/neuron-agentic-development/neuron-framework-autoport-vllm-neuron)
Your own site
<a href="https://agentmods.dev/skills/aws-neuron/neuron-agentic-development/neuron-framework-autoport-vllm-neuron"><img src="https://agentmods.dev/badge/skills/aws-neuron/neuron-agentic-development/neuron-framework-autoport-vllm-neuron.svg" alt="Measured on agentmods" height="20"></a>
Per session 40 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,847 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. Scan, not verified.
Origin original 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 5.1 $0.00040 $0.07847
Opus 5 $0.00020 $0.03923
Sonnet 5 $0.00008 $0.01569
Haiku 4.5 $0.00004 $0.00785

Measured 6d ago against content hash 06118c2aa752, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade B, and why

neuron-framework-autoport-vllm-neuron scanned grade B with 1 finding 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 6d 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.

Unrestricted tool accessmediumExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

- **Do not run any code** — no compilation, inference, or validation (no Trainium hardware available)
skills/neuron-framework-autoport-vllm-neuron/SKILL.md · 507 lines

How it starts

The opening of the file, as written. The whole thing — 507 lines — stays where its author put it; the contents beside it link to each section on GitHub.

vLLM-Neuron Model Porting Skill

Port the HuggingFace model $1 to the vLLM-Neuron Trainium2 backend as $0.

Conventions

  • MODEL_NAME = $0 (snake_case, e.g., yi, deepseek_v2)
  • HF_MODEL_ID = $1 (e.g., 01-ai/Yi-6B-Chat)
  • --review flag (if present in $ARGUMENTS): pause at Step 2 for user confirmation
  • Derive PascalName from MODEL_NAME (e.g., yiYi, gpt_neoxGPTNeoX)
  • Model code goes in vllm_neuron/model
  • Example script goes in examples/vllm_neuron/models/MODEL_NAME/

Dry-run

When the user specifies dry-run:

  • Skip all agent-level prerequisites (package import checks, NeuronCore check via neuron-ls)
  • Activate the provided venv and resolve source paths by filesystem lookup (do NOT use import — dependencies are not installed):
    export PATH=<pathToVenv>/bin:$PATH
    VLLM_NEURON_SRC=~/tmp/vllm_autoport/vllm_neuron
    TRANSFORMERS_SRC=$(python3 -c "import site; print(site.getsitepackages()[0])")/transformers
    
  • Do not run any code — no compilation, inference, or validation (no Trainium hardware available)

Before You Start

Ensure the agent-level prerequisites have been completed (venv activation, package verification, NeuronCore check). These steps are owned by the agent prompt and should not be re-executed here.

Note: Do NOT clear /var/tmp/neuron-compile-cache as a pre-flight step — it is a shared system directory and other processes or users may depend on it. Only clear it reactively if you hit a [NLA001] JSON parse error or FileNotFoundError on neff_output paths.

Step 0: Parse & Detect State

  1. Package check: Ensure that private_vllm_neuron or vllm_neuron package exists in the user's workspace. If it does not, STOP and do not proceed without informing the user.
  2. Parse $ARGUMENTS — extract model-name, hf-model-id, and --review flag.
  3. Check if vllm_neuron/model/MODEL_NAME/ already exists:
    • If model.py + registry entry both exist → skip to Phase C (validation)
    • If model.py exists but not registered → resume at Step 5 (register)
    • If directory doesn't exist → proceed to Phase A

Read the full file on GitHub · 507 lines

Files

What ships with it

6 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. 6d ago First seen · 507 lines · 40 tokens per session scan B 06118c2aa752

Subscribe to this mod's changes

neuron-framework-autoport-vllm-neuron is a skill published in the GitHub repository aws-neuron/neuron-agentic-development (57 stars, last pushed 17d ago), licensed Apache-2.0. It adds 40 tokens to every session and 7,847 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

spark-environment-setup

Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.

wshobson/agents · 76 tokens

spark-memory-thermal-ops

Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.

wshobson/agents · 59 tokens

spark-training-gotchas

Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.

wshobson/agents · 63 tokens

llama-cpp

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.

davila7/claude-code-templates · 76 tokens

amc-run-rtsp-calibration

Calibrate a new dataset from live RTSP camera streams via the AutoMagicCalib REST API. Use when the user provides RTSP URLs or asks to calibrate live cameras; VIOS records clips, AMC ingests them, then runs calibration.

NVIDIA/skills · 59 tokens

amc-run-video-calibration

Calibrates pre-recorded cam.mp4 datasets through the AutoMagicCalib REST API. Use for user-supplied local MP4s; route live RTSP streams to amc-run-rtsp-calibration.

NVIDIA/skills · 57 tokens