Reference Board logo

Reference Board

Community
machina-exm
reference-board

Build and lock reference boards before asset creation.

Overview

Publishermachina-exm
Repositoryfilm-studio-skills
Skill namereference-board
Stars
156
Forks
25
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by machina-exm on GitHub. Read the source before you install it.

Installation

Install the Reference Board AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/machina-exm/film-studio-skills/tree/main/skills/reference-board
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/machina-exm/film-studio-skills.git /tmp/film-studio-skills
mkdir -p .claude/skills
cp -r /tmp/film-studio-skills/skills/reference-board .claude/skills/reference-board
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Reference Board in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Reference Board on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Reference Board is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Reference Board

Turn references supplied by the user into a written visual bible. Treat references as specifications, not inspiration.

Locate the studio

Work in /docs/bible.md under the current studio root. Read /docs/breakdown.md and /docs/registry.md when present to collect known assets, locations, props, scenes, and tags. If /docs/ or /docs/bible.md is missing, offer to create the needed folder or document. Ask one question at a time and wait for the answer.

Set the reference rule

First find an existing image, then describe it. Never start with an imagined description and search for an image to justify it. Do not fetch or generate images. Ask the user to supply each reference as a file, link, or other stable identifier. Help the user describe and organize what they supply.

Build boards on two axes

Create asset boards and style-category boards in /docs/bible.md.

Axis 1: per asset

Use these target ranges:

  • lead character: 10–20 images
  • key location: 8–15 images
  • prop: 3–5 images

Create one board for each asset. Keep asset tags and state variants distinct. Do not combine separate assets into one board.

Axis 2: per style category

Create separate boards for:

  • light
  • color
  • optics
  • camera movement
  • texture
  • cutting tempo
  • sound

Do not invent a target count where none is supplied.

Interview one board at a time

Choose one asset or style category with the user. Ask for one existing reference. Then ask what exact element should be taken from that reference. Continue one question at a time until the board is ready for a decision. Move to the next board only after recording the current board's decision.

Write every reference entry

For each supplied reference, record:

FieldRequired content
ReferenceFile, link, or stable identifier
KindReference or anti-reference
BoardAsset tag or style category
CaptionExactly what is taken from it

Write a precise caption for every image. Reject an image that the user only "just likes" until they can name what it specifies. Do not infer a caption without confirming it with the user.

Build the ban list

Mark anti-references as like this — not allowed. For each anti-reference, name exactly which visible or audible property is forbidden. Copy confirmed anti-reference rules into the ## Ban list section of /docs/bible.md. Keep the source reference attached to each ban.

Lock every board in writing

End each asset and style board with one fixed written decision:

  • approved
  • revise
  • rejected

Record the decision on the board itself. Treat verbal approval as no approval until the written decision is present. If the decision is revise, record the exact unresolved issue and continue one question at a time. Do not describe a board as locked unless its written decision is approved.

Run the lock checklist

Before reporting completion, verify:

  • every lead character has 10–20 supplied images
  • every key location has 8–15 supplied images
  • every prop has 3–5 supplied images
  • every supplied image has a specific caption
  • every anti-reference appears in the ban list
  • every asset and style board has a written decision
  • only boards marked approved are called locked

List missing references, captions, bans, or decisions without filling them in. Resolve gaps by asking one question at a time.

Report the result

State which boards are approved, need revision, or are rejected. Name the exact sections updated in /docs/bible.md. Remind the user that the next stage may rely only on the written, approved boards.

Frequently asked questions

What does the Reference Board AI skill do?

Build and lock reference boards before asset creation.

Why use Reference Board on TypingMind?

Because you install it once and use it with any model. Reference Board is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Reference Board in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/machina-exm/film-studio-skills/tree/main/skills/reference-board. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Reference Board?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Reference Board?

As many as you like. As long as a model supports skills, you can use Reference Board with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Reference Board AI skill free?

It is published on GitHub by machina-exm. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇