Colormason

Community

vladzima

colormason

Generates a complete color system from a single brand color - perceptually even OKLCH ramps (50-950) plus WCAG-checked semantic design tokens (backgrounds, text, borders, focus ring) with light/dark values - via a zero-dependency local Python script. Use when starting or restyling a project's colors, generating design tokens or a Tailwind theme from a brand hex, building a light/dark palette, picking accessible UI colors, or when the user mentions color palette, design tokens, theme colors, brand color, OKLCH, or WCAG color contrast.

Overview

Publisher	vladzima
Repository	`colormason`
Skill name	`colormason`
Stars	3
Forks	0
Bundled files	3
License	MIT
Links	Skill source Repository Homepage

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.
3 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by vladzima on GitHub. Read the source before you install it.

Installation

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

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.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/vladzima/colormason/tree/main/colormason
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.

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/vladzima/colormason.git /tmp/colormason
mkdir -p .claude/skills
cp -r /tmp/colormason/colormason .claude/skills/colormason

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Colormason 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 Colormason 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 Colormason 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.

colormason

Turn one brand hex into a full, consistent, accessible color system: 8 ramps (11 steps each) and ~45 semantic tokens with light/dark values, emitted as CSS variables, Tailwind v4 theme, JSON, or Markdown. Fully deterministic and offline - never invent palette hex values by hand when this skill is available.

When to use

New project or rebrand needs a color system / theme file
User gives a brand color and asks for a palette, tokens, or theme
Building or fixing light/dark mode colors
Tailwind @theme, CSS variables, or design-token file needs generating
Checking or repairing color contrast (WCAG AA/AAA) in a token set

Do not use for one-off decorative colors (a single gradient, an illustration), for pinning colors that must match an existing system exactly (use that system's values verbatim instead), or when the project already has a token file - extend it, don't generate a competing one.

Inputs to collect first

Brand hex (required). Ask the user, or extract from a logo / existing CSS. Any hex works; near-neutral brands (grays) produce a warning because accent hues become arbitrary - suggest pinning accents in that case.
Scheme (default complementary): complementary, analogous, triadic, split, monochromatic. Pick analogous for calm/conservative brands, triadic for playful, complementary for max accent distinction.
Scope (default full): basic drops the accent ramps (keeps primary, neutral, status colors).
WCAG target (default AA): use AAA for text-heavy or government/health products.
Pins (optional): exact --accent / --accent-2 hexes when the brand system already specifies secondary colors.

If the user hasn't decided, do not interview them - run the defaults and offer the knobs after showing output.

Procedure

Run the generator (script is relative to this skill's directory):

bash
python3 scripts/palette.py --brand 3d7dff --format css

Pick the format for the target project:
- css - plain CSS custom properties (default; works everywhere)
- tailwind - Tailwind v4 @theme + @theme inline blocks with light/dark variables
- json - machine-readable; use when feeding a token pipeline
- markdown - human-readable summary for review/discussion
Write the output to the project (e.g. src/theme.css or into the main CSS file / Tailwind entry). Do not rename tokens downstream - other tools and agents may rely on the exact contract in references/tokens.md.
Report to the user: brand, scheme, target, any warnings from stderr, and the two or three token pairs that were auto-adjusted (visible in JSON as contrast.*.adjusted: true).

Rules

Use semantic tokens (--bg-canvas, --text-primary, ...) in UI code, not raw ramp steps. Raw steps bypass the light/dark mapping and the contrast checking, so components break in dark mode or fail WCAG.
Never hand-edit generated hex values. Each foreground/background pair is contrast-verified; a "small tweak" silently breaks the verified ratio. Regenerate with different inputs instead (scheme, pins, target).
The brand hex will not appear verbatim in the ramps. Steps sit on a fixed OKLCH lightness ladder so the scale stays perceptually even. This is expected - use the ramp values, not the original input. If the user insists on exact brand reproduction, primary-500 is the nearest step.
bg-surface-raised equals bg-surface in light mode by design - in light mode, elevation must read through shadow, not tone. Add a shadow there instead of "fixing" the token.
text-disabled is intentionally below WCAG minimums (disabled controls are exempt). The exemption does not extend to real text - use text-tertiary for low-emphasis but readable copy.
Warnings mean what they say. If a pair cannot reach the target at any ramp step (common for AAA on colored buttons), either accept it, lower the target for that pair, or pin a darker accent. Do not silence it.

Edge cases

Gray/neutral brand: script warns; recommend --accent/--accent-2 pins.
Existing brand has secondary colors: pass them as pins so the palette matches instead of deriving.
User wants fewer ramps (no info color, etc.): --exclude info also drops every token that references the excluded ramp.
AAA on white-on-brand labels is physically unreachable on most hues; the script keeps the best step and emits a warning rather than breaking the design.

References

references/tokens.md - full token contract: every token name, ramp, light/dark steps, role. Read this when writing components against a colormason palette.
references/integration.md - wiring output into Tailwind v3/v4, plain CSS, CSS-in-JS, and shadcn-style variable themes.
python3 scripts/palette.py --help - all options.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Colormason AI skill do?

Why use Colormason on TypingMind?

Because you install it once and use it with any model. Colormason 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 Colormason in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vladzima/colormason/tree/main/colormason. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Colormason?

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 Colormason?

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

Is the Colormason AI skill free?

Yes. It is published on GitHub by vladzima 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

Sketch

NousResearch

OrganizationPopular

Throwaway HTML mockups: 2-3 design variants to compare.

Tldraw Offline

Drive and script tldraw offline canvases with an agent.

Brand Guidelines

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual…

Solid

Write, port, review, or test Solid 2 UI in OpenClaw (Control UI, plugin views, Canvas hosts), including state ownership, Lit interop during the migration, native overlays, testing, performance, and…

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

colormason

When to use

Inputs to collect first

Procedure

Rules

Edge cases

References

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now