training_hub: Skill for Claude Code

.claude/skills/setup-guide/SKILL.md

setup-guide is a skill for Claude Code from Red-Hat-AI-Innovation-Team/training_hub. It costs 35 tokens per session (988 once invoked), scanned A, original, Apache-2.0.

A first-time setup procedure for Training Hub, the software used to fine-tune language models.

In plain words
What is it for?
Use it to install Training Hub, choose optional support for CUDA, LoRA, or GRPO training, detect available GPUs, and prepare the environment before training.
Why use it?
It checks the environment, installs the base package and optional training components, and warns when no CUDA-capable GPU is available.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

This is Red-Hat-AI-Innovation-Team/training_hub's own configuration. It tells Claude Code how to work on training_hub itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything training_hub configures →

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the training-hub plugin — 4 skills shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to Red-Hat-AI-Innovation-Team/training_hub. 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/Red-Hat-AI-Innovation-Team/training_hub/main/.claude/skills/setup-guide/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Red-Hat-AI-Innovation-Team/training_hub

Made for: Claude Code.

Or install training-hub, the plugin that ships this one along with the rest of its 4 skills.

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 setup-guide

README.md
[![agentmods](https://agentmods.dev/badge/skills/red-hat-ai-innovation-team/training_hub/setup-guide/github.svg)](https://agentmods.dev/skills/red-hat-ai-innovation-team/training_hub/setup-guide)
Your own site
<a href="https://agentmods.dev/skills/red-hat-ai-innovation-team/training_hub/setup-guide"><img src="https://agentmods.dev/badge/skills/red-hat-ai-innovation-team/training_hub/setup-guide/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.

agentmods 80×15 button for setup-guide

Your own site · 80×15
<a href="https://agentmods.dev/skills/red-hat-ai-innovation-team/training_hub/setup-guide"><img src="https://agentmods.dev/badge/skills/red-hat-ai-innovation-team/training_hub/setup-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 988 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 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 $0.00035 $0.00988
Opus 5 $0.00017 $0.00494
Sonnet 5 $0.00007 $0.00198
Haiku 4.5 $0.00003 $0.00099

Measured 10d ago against content hash 5e51a6f1570d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

setup-guide 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 10d 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/setup-guide/SKILL.md · 103 lines

How it starts

The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.

training_hub Setup Guide

You are helping the user set up LLM training. For algorithm selection guidance, hyperparameter tuning, and troubleshooting, consult the training-hub-guide skill.

Step 1: Detect Environment

"${CLAUDE_PLUGIN_ROOT}/scripts/th_detect.sh"

Step 2: Install if Needed

If library=missing:

  • Ask permission: "training_hub isn't installed. I can install it for you — want me to proceed?"
  • If yes and installer=uv: run uv pip install training-hub
  • If yes and installer=pip: run pip install training-hub
  • If installer=none: tell the user they need Python and pip/uv installed first
  • Ask about extras:
    • CUDA: training-hub[cuda] — flash-attn, bitsandbytes for GPU acceleration
    • LoRA: training-hub[lora] — Unsloth, TRL for parameter-efficient fine-tuning
    • GRPO: training-hub[grpo] — ART, veRL for reinforcement learning

For installation issues, consult the training-hub-guide skill (installation-troubleshooting section).

Step 3: Check GPU

If gpu=unavailable, warn: "No GPU detected. Training requires CUDA-capable GPUs. You can still configure, but training will fail without a GPU."

Report GPU count if available.

Step 4: Quick Setup or Custom

If the user has a clear task ("fine-tune Llama on my data"), offer a fast path with sensible defaults:

"I detected N GPU(s). I can set up with these defaults:

  • Algorithm: lora_sft (parameter-efficient, works on a single GPU)
  • Learning rate: 1e-5
  • Epochs: 2
  • Batch size: 64
  • Max sequence length: 4096

You'll just need to provide your model path and data path. Accept these defaults, or customize?"

If the user accepts, ask only for model path and data path, then skip to Step 7.

If the user wants to customize, proceed with the full configuration.

Full Configuration

Ask these questions one at a time:

  1. Algorithm: "Which training algorithm do you want to use?" — consult the training-hub-guide skill for algorithm selection guidance if the user is unsure.
  2. Model path: "What's the model identifier?" — e.g., meta-llama/Llama-3.1-8B-Instruct, or a local path.
  3. Data path: "Where is your training data?" — Path to a JSONL file with messages field.
  4. Output directory: "Where should checkpoints be saved?" — Default: ./output
  5. GPU count: "How many GPUs should be used?" — Default: detected count or 1.

Read the full file on GitHub · 103 lines

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. 10d ago First seen · 103 lines · 35 tokens per session scan A 5e51a6f1570d

Subscribe to this mod's changes

setup-guide is a skill published in the GitHub repository Red-Hat-AI-Innovation-Team/training_hub (95 stars, last pushed today), licensed Apache-2.0. It adds 35 tokens to every session and 988 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

ml-training-recipes

Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning…

Orchestra-Research/AI-Research-SKILLs · 88 tokens

tao-run-automl

Run container-backed AutoML / hyperparameter optimization (HPO) for NVIDIA TAO networks using AutoMLRunner. Handles algorithm selection (bayesian, hyperband, asha, bohb, llm, hybrid, autoresearch), WandB experiment tracking, job execution on any TAO SDK platform, result interpretation, and per-rec custom evaluation…

NVIDIA-TAO/tao-skill-bank · 235 tokens

unsloth

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization.

davila7/claude-code-templates · 39 tokens

unsloth

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization.

OpenLAIR/dr-claw · 39 tokens

hugging-face-model-trainer

This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV…

synthetic-sciences/openscience · 131 tokens

Training

Guides fine-tuning and post-training work with explicit data, objective, hardware, and rollback assumptions.

agentic-in/elephant-agent · 23 tokens