Layer Reference Sets

Use when a look must hold across many Layer generations: training a custom style or LoRA on a studio's own artwork, curating the images that go into a reference set, choosing its training category, tuning reference-set weight on a run, or deciding whether to train at all rather than attach a style reference. Also when a trained style produces weak, inconsistent, or silently ignored results. Keywords: LoRA, custom model, trained style, reference set, consistency, on-model, art direction, dataset.

10 stars 3 forksAI, LLMs & Agents Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add layerai/skills

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add layerai/skills --skill layer-reference-sets
Manual install
git clone --depth 1 https://github.com/layerai/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/plugins/layer/skills/layer-reference-sets ~/.claude/skills/layer-reference-sets

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Layer Reference Sets

## Overview

A reference set is a named group of files plus a training category. Trained, it produces a LoRA that
later runs apply by id, turning an art direction into a parameter instead of a prompt-engineering
exercise every time.

The mechanics (upload, `create_reference_set`, `estimate_training_price`, `start_training`,
`get_training_status`, `get_trained_model`, and the four categories) are covered by Layer's published
`layer-train-reference-set` skill, over both REST and MCP. This skill covers the decisions that
determine whether the trained result is any good.

If a sibling skill named here is missing from your available skills, ask the user to install it
(`npx skills add layerai/skills --skill <name>`); unattended, proceed from tool schemas and flag the
gap.

## Decide whether to train at all

Training costs Creative Units and takes time. It is the right answer less often than people expect.

| Situation                                   | Do this                                                                      |
| ------------------------------------------- | ---------------------------------------------------------------------------- |
| A handful of assets, this week              | Attach an approved image as `reference_image`, filtering for `image_editing` |
| One character across a dozen images         | An approved hero image as a reference usually suffices                       |
Read the full skill on GitHub

Details

Skill name
layer-reference-sets
Repository
layerai/skills
Path
plugins/layer/skills/layer-reference-sets/SKILL.md
License
MIT
Repo updated
18 November 2026

FAQ

What does the Layer Reference Sets skill do?

Use when a look must hold across many Layer generations: training a custom style or LoRA on a studio's own artwork, curating the images that go into a reference set, choosing its training category, tuning reference-set weight on a run, or deciding whether to train at all rather than attach a style reference. Also when a trained style produces weak, inconsistent, or silently ignored results. Keywords: LoRA, custom model, trained style, reference set, consistency, on-model, art direction, dataset.

How do I install the Layer Reference Sets skill?

Claude Code plugin: /plugin marketplace add layerai/skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Layer Reference Sets popular?

It lives in layerai/skills, which has 10 GitHub stars and 3 forks.

Related skills

Layer

10

layerai/skills

Use when generating or editing game and entertainment assets on Layer through MCP: images, video, 3D meshes, audio, sprites, textures, or marketing creative. Also when choosing a Layer base model, pricing a run in Creative Units before spending them, uploading reference files, polling a generation, applying a trained style, or recovering from MODEL_NOT_FOUND, INSUFFICIENT_BALANCE, RATE_LIMITED, or a silently degraded reference set.

Images, Video & Audio

Layer 3d

10

layerai/skills

Use when producing 3D assets with Layer: generating a mesh from text or an image, retopologising or remeshing, retexturing an existing mesh, rigging a character, or animating a rigged mesh. Also when choosing between quad and triangle output, setting a polygon budget, or asking for PBR textures for a game engine. Keywords: 3D, mesh, model, text to 3D, image to 3D, remesh, retopology, retexture, rig, skeleton, animate, PBR, quad mesh, poly count, glTF, FBX.

Images, Video & Audio

Layer Art Direction

10

layerai/skills

Use when a Layer brief needs a direction before generating at volume: exploring a look with a moodboard, choosing between visual directions, storyboarding an ad, trailer or cutscene, pinning a colour palette, naming a style or era precisely, or framing a scene. Also when a brief is vague or generations keep missing an unstated intent. Keywords: art direction, moodboard, style exploration, storyboard, panels, beats, palette, composition, framing, era, reference.

Writing & Content

Octo Agent

10

meshmakers/octo-claude-skills

Debugging and investigation agent for OctoMesh — diagnoses CK model failures (ResolveFailed, modelState 2, broken dependencies, version-bump cascades), blueprint install failures, and CK migration problems; inspects MongoDB tenant databases directly; runs selective DebugL builds and targeted tests to isolate a regression; and manages Docker volume backups for safe rollback. Use it when something broke after a build, version bump, merge, or upgrade and you need to find the root cause safely. Trigger on: help me fix, investigate this bug, what broke, isolate the issue, why is the model broken, R

AI, LLMs & Agents

Layer Game Assets

10

layerai/skills

Use when producing in-game characters and props with Layer: heroes, NPCs, enemies, bosses, mascots, creatures, items and props. Also when a cast must stay on-model across many images, when a set of sprites must share one canvas size and ground line, or when a character must be split into parts for a skeletal rig. Keywords: character art, NPC, enemy, mascot, cast, on-model, consistency, prop, item, sprite set, Spine, rigging.

AI, LLMs & Agents

Octo Logs

10

meshmakers/octo-claude-skills

Query and trace OctoMesh logs, traces and metrics through the Dash0 MCP server — no credentials, port-forwards or logcli needed. Filter by namespace/container/pod/service/severity, search message bodies, count error rates with D0QL (SQL), trace an error across pod redeployments and jump from a log line to its trace. Loki/Promtail/LogQL are not used any more. Run /octo-logs-setup if the dash0 MCP tools are missing. Trigger on: logs, cluster logs, view logs, service logs, error logs, grep logs, tail logs, Dash0 logs, getLogRecords, D0QL, trace error, error over deployments, log retention, namesp

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.