Vellum Workspace Theme logo

Vellum Workspace Theme

OrganizationPopular
vellum-ai
vellum-workspace-theme

Customize the app's visual theme — author design-token overrides (accent, background, surfaces, text, message-bubble colors) in the workspace ui/theme.json, validated by the assistant runtime and applied live to connected clients. Covers the token slots, the all-or-none override groups, the contrast floor, and reading back validation issues.

Overview

Publishervellum-ai
Repositoryvellum-assistant
Skill namevellum-workspace-theme
Stars
1.4K
Forks
208
Bundled files
Instructions only
LicenseMIT
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 vellum-ai on GitHub. Read the source before you install it.

Installation

Install the Vellum Workspace Theme 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/vellum-ai/vellum-assistant/tree/main/skills/vellum-workspace-theme
  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/vellum-ai/vellum-assistant.git /tmp/vellum-assistant
mkdir -p .claude/skills
cp -r /tmp/vellum-assistant/skills/vellum-workspace-theme .claude/skills/vellum-workspace-theme
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Vellum Workspace Theme 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 Vellum Workspace Theme 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 Vellum Workspace Theme 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.

You are customizing the visual theme of the app you and your user share. The theme lives in one file — $VELLUM_WORKSPACE_DIR/ui/theme.json — which the assistant runtime validates and serves to every connected client. Valid changes apply live, layered on top of the built-in light/dark/velvet base theme: open windows re-color within a few seconds of the file being saved. No restart, no build step.

All commands in this skill use the bash tool. $VELLUM_WORKSPACE_DIR is available in the sandbox environment.

Inspect current state

Always check what's there before changing it:

bash
cat "$VELLUM_WORKSPACE_DIR/ui/theme.json" 2>/dev/null || echo "No theme yet — clients render the built-in theme"

The file

A complete example (all eleven token slots — you rarely need all of them):

json
{
  "version": 1,
  "tokens": {
    "accent": "#e8a04c",
    "background": "#1c1512",
    "surface": "#2b2018",
    "surfaceRaised": "#332619",
    "border": "#43301f",
    "text": "#f2e4d4",
    "textMuted": "#a68d75",
    "userBubbleBackground": "#26201a",
    "userBubbleText": "#efe3d2",
    "assistantBubbleBackground": "#33202a",
    "assistantBubbleText": "#ffd7e4"
  }
}
TokenWhat it controls
accentPrimary buttons, highlights, active states. The button-label color is derived automatically so it stays readable on any accent.
backgroundThe page background and base surfaces.
surfacePanels, popovers, and overlays.
surfaceRaisedElevated cards.
borderBorders and dividers.
textMain copy.
textMutedSecondary labels and captions.
userBubbleBackground / userBubbleTextThe user's chat message bubble.
assistantBubbleBackground / assistantBubbleTextAccepted and validated, but not yet rendered — assistant messages don't have a themeable container yet. Safe to set now; they light up when the surface ships.

"base": "light" | "dark" | "velvet" | "system" is also accepted at the top level but reserved — clients don't switch the base theme from it yet. The tokens are a color layer over whichever base the user has chosen.

The rules (why an edit gets rejected)

The runtime validates the whole file and rejects it atomically — one bad value turns theming off (clients revert to the built-in theme) until the file is fixed. The rules:

  1. Hex colors only — 3- or 6-digit (#f0a, #e8a04c). No alpha channel, no named colors, no CSS functions, no url().
  2. Only the keys above — unknown top-level keys or token slots reject the file. There is no way to inject arbitrary CSS; that is deliberate.
  3. Override groups are all-or-none. Tokens that render against each other must be set together, because your overrides layer over a base palette the runtime doesn't resolve:
    • Core group: background, surface, surfaceRaised, text, textMuted — set any one, set all five.
    • Each bubble pair: background and text together.
    • accent and border are free agents — fine alone.
  4. Contrast floor — text/surface pairs must clear 3:1 (WCAG relative luminance): text against background/surface/surfaceRaised, textMuted against all three, and each bubble's text against its background. The floor blocks illegible themes, not bold ones.
  5. A real file, at most 64KB — symlinks and oversized files are rejected.

Practical consequence of rule 3: an "accent-only" theme is two lines, but the moment you touch background you are authoring a full palette. Design the five core colors together.

Verify after every edit

The runtime tells you exactly what it accepted or why it refused. Fetch GET /v1/workspace/theme from the assistant runtime HTTP API (the same authenticated API used for other configuration reads):

json
{
  "theme": { "version": 1, "tokens": { "accent": "#e8a04c" } },
  "source": "workspace",
  "issues": []
}
  • source: "workspace" — your theme is live.
  • source: "invalid" — rejected; issues lists human-readable reasons (e.g. "incomplete override group: setting tokens.background also requires tokens.text, …" or "tokens.text on tokens.background has contrast 1.02:1 — minimum is 3:1"). Fix and re-check.
  • source: "none" — no theme file exists.

Seeing your work

You do not need the user to screenshot the app for you. The desktop app can render a staged view of its own UI — fixed generic content, never real conversation data — with your current theme applied, capture it offscreen, and hand you the PNG:

bash
assistant ui snapshot --view sampler --out /tmp/theme-sampler.png
assistant ui snapshot --view chat --out /tmp/theme-chat.png

Then view the image file. Two views:

  • sampler — a dense style sheet: the text ramp, accent, buttons, a raised card, inputs, borders, and both chat bubbles in one frame. Use it while iterating — it answers "does the palette read".
  • chat — a staged conversation with a composer. Use it to judge the finished feel — it answers "does it look like home".

The capture reflects ui/theme.json as it is on disk at that moment, so edit → snapshot → look → adjust is a tight private loop. If the theme file is invalid, the command prints the validation issues and the capture shows the built-in theme. The framing is fixed, so before/after snapshots align exactly.

This requires the desktop app to be running; if the command times out or reports no connected client, fall back to taking a browser screenshot of the app or asking the user — the change is already on their screen.

Removing the theme

Delete the file (or write {"version":1} with no tokens) and clients revert to the built-in theme:

bash
rm "$VELLUM_WORKSPACE_DIR/ui/theme.json"

Etiquette

The theme applies to every open window your user has, within seconds, with a small "Theme updated" notice. For a small tweak that was just discussed, go ahead. For a dramatic repaint — or one the user didn't ask for — tell them what you're about to do, or stage the palette in conversation first. It's a shared space; redecorate like someone who lives there too.

Frequently asked questions

What does the Vellum Workspace Theme AI skill do?

Customize the app's visual theme — author design-token overrides (accent, background, surfaces, text, message-bubble colors) in the workspace ui/theme.json, validated by the assistant runtime and applied live to connected clients. Covers the token slots, the all-or-none override groups, the contrast floor, and reading back validation issues.

Why use Vellum Workspace Theme on TypingMind?

Because you install it once and use it with any model. Vellum Workspace Theme 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 Vellum Workspace Theme in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vellum-ai/vellum-assistant/tree/main/skills/vellum-workspace-theme. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Vellum Workspace Theme?

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 Vellum Workspace Theme?

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

Is the Vellum Workspace Theme AI skill free?

Yes. It is published on GitHub by vellum-ai under the MIT license. 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 👇