// Use when the user wants to run a training job using a saved configuration. For algorithm selection, hyperparameter advice, or troubleshooting, use the training-hub-guide skill instead.
Use when the user wants to estimate GPU memory (VRAM) requirements for a training configuration, check if a model will fit on their GPUs, or plan GPU allocation for training.
Use when the user wants to set up LLM training for the first time, or when training_hub is not yet installed/configured in the current environment.
Guides users through LLM post-training with Training Hub, including installation, algorithm selection (SFT, OSFT, LoRA), hyperparameter tuning, troubleshooting OOM errors, interpreting loss curves, and leveraging backend-specific features. Use when the user is working with training_hub, fine-tuning language models, asking about SFT/OSFT/LoRA training, or debugging GPU/CUDA training issues.
| name | training-guide |
| description | Use when the user wants to run a training job using a saved configuration. For algorithm selection, hyperparameter advice, or troubleshooting, use the training-hub-guide skill instead. |
| allowed-tools | ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/th_train.sh:*)","Bash(${CLAUDE_PLUGIN_ROOT}/scripts/th_detect.sh:*)"] |
Execute LLM training using a saved configuration.
"${CLAUDE_PLUGIN_ROOT}/scripts/th_detect.sh"
library=missing or config=missing: invoke the setup-guide skill.gpu=unavailable: warn that training requires CUDA-capable GPUs.library=installed, config=found)Proceed to Step 2.
Run the training script with any user-provided overrides:
"${CLAUDE_PLUGIN_ROOT}/scripts/th_train.sh" $ARGUMENTS
If training failed, consult the training-hub-guide skill for troubleshooting (OOM, loss interpretation, backend-specific issues).
Remind the user they can visualize training loss with training_hub.plot_loss().