train-lora — Claude Code skill for LoRA training checks
train-lora is a Claude Code skill that validates LoRA training with Draw Things through draw-things-cli. It runs local dataset checks and compares generated images with and without the LoRA.
GPL-3.0 ★ 171 updated 2026-11-17
npx skills add drawthingsai/local-code-community --skill train-lora -a claude-codeWhat train-lora does
train-lora checks LoRA training from a short trial run through image comparison. First, build the optimized CLI with bazel build --compilation_mode=opt //Apps:DrawThingsCLI, then run bazel-bin/Apps/DrawThingsCLI for the training and generation commands.
A single-image test uses a local dataset folder containing the image and a matching .txt caption file. The workflow checks runs at 1, 20, 100, and 500 steps, tracking loss, scale, and checkpoint output. For generation checks, it holds the prompt, seed, and settings constant between the base model and the LoRA. Graph compilation in optimized mode can take 5 to 30 minutes on heavy trainers.
What it helps with
- check graph compilation with a short training run
- track loss and scale during training
- verify that a checkpoint contains nonzero LoRA weights
- compare base and LoRA images under matching settings
Example prompts for the agent
- Validate Draw Things LoRA training with a small local dataset.
- Check loss, scale, and checkpoint output during a LoRA run.
- Compare base-model and LoRA images using identical generation settings.
How to install train-lora
1Skills CLI (Vercel)
npx skills add drawthingsai/local-code-community --skill train-lora -a claude-code2Manually: put the files in place
~/.claude/skills/train-lora/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/train-lora/
1Skills CLI (Vercel)
npx skills add drawthingsai/local-code-community --skill train-lora -a codex2Manually: put the files in place
.agents/skills/train-lora/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add drawthingsai/local-code-community --skill train-lora -a cursor2Manually: put the files in place
.cursor/skills/train-lora/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install drawthingsai/local-code-community .agents/skills/train-lora2Manually: put the files in place
.github/skills/train-lora/SKILL.md1Skills CLI (Vercel)
npx skills add drawthingsai/local-code-community --skill train-lora -a gemini-cli2Platform CLI
gemini skills install https://github.com/drawthingsai/local-code-community --path .agents/skills/train-loraManually: put the files in place
.devin/skills/train-lora/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about train-lora
How do I install train-lora in Claude Code?
Run npx skills add drawthingsai/local-code-community --skill train-lora -a claude-code. The skill file is placed at ~/.claude/skills/train-lora/SKILL.md.
What does train-lora do?
It validates Draw Things LoRA training through staged runs, checks loss, scale, and checkpoint output, then compares generated images from the base model and LoRA.
What is required to use train-lora?
It requires draw-things-cli, built with Bazel, a local model, and an image dataset. For a single-image check, the image needs a matching .txt caption file.