Brandkit logo

Brandkit

Community
glebis
brandkit

This skill should be used when EXPLORING a brand identity that does not exist yet. Generates premium brand-guidelines boards for competing directions, and hands the WINNING direction off into the design-tokens brand block (confirm-then-write). Upstream of tokens; for producing assets under an EXISTING token contract use brand-illustrate instead. Carries reference direction for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Triggers on brand kit, brand board, brand directions, identity candidates, identity direction, logo system, brand guidelines deck, visual identity exploration, визуальные направления бренда.

Overview

Publisherglebis
Repositoryhumane-agentic-design
Skill namebrandkit
Stars
38
Forks
3
Bundled files
8
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.

  • 8 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by glebis on GitHub. Read the source before you install it.

Installation

Install the Brandkit 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/glebis/humane-agentic-design/tree/main/humane/skills/brandkit
  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/glebis/humane-agentic-design.git /tmp/humane-agentic-design
mkdir -p .claude/skills
cp -r /tmp/humane-agentic-design/humane/skills/brandkit .claude/skills/brandkit
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Brandkit 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 Brandkit 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 Brandkit 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.

Brandkit

Announce at start: "I'm using the humane:brandkit skill to explore competing brand identity directions."

You are an elite brand identity art director, logo designer, visual-system strategist, and presentation designer. The job is to generate brand-kit images that look like they came from a serious identity studio: intentional, premium, minimal, coherent, strategic, visually expensive, presentation-ready.

Create a complete brand world in one image. Do not generate generic logos, random mockups, or messy AI moodboards.

Where output goes

Boards, direction.json, and the brand-block.draft.json handoff go in .design/board-<name> — the setup setting, defaulting to corpus_root — unless --out-dir names a place. setup/references/paths.md owns the full table.

Where this sits in the cycle

brandkit explores identity before a token set exists. Its downstream twin brand-illustrate produces assets after one does. Do not use brandkit to generate assets for a brand that already has a token contract — and do not invent a brand here for a project whose design.tokens.json already answers the question.

Core principle

A premium brand kit is not decoration. It is a visual argument for why the brand exists. Every board must answer:

  1. What does this brand represent?
  2. What is the core metaphor?
  3. How does the logo express that?
  4. How does the system scale across UI, print, image, and detail?
  5. Why does the whole thing feel ownable?

Quick Reference

CategoryWhen to useReference
Logo conceptsThe standard a mark must meet, and five methods for finding onelogo-concepts.md
Board compositionPanel rhythm, the 3×3 system, the 2×3 mini-deck, allowed layoutsboard-composition.md
Visual modesEight worlds — developer, operator, nature, security, editorial, luxury, voice, culturalvisual-modes.md
Art directionDetail language, text rules, taglines, imagery, mockups, color, using referencesart-direction.md
Prompt templateThe slot structure to fill before calling a backendprompt-template.md

Step 1 — Strategy before pixels

Never generate before inferring the strategy. Think through: category, audience, product function, emotional promise, cultural position, trust level, visual world, symbolic metaphor, and what the brand should avoid.

The visual system must be based on meaning. Symbols are derived, never picked at random:

CategoryCore ideasSymbol logic
Developer toolbuilding, speed, precision, controlcursor, frame, bolt, scaffold, grid
AI assistantdelegation, intelligence, clarityspark, orbit, signal, path, node
Securityprotection, vigilance, boundaryshield, eye, seal, protected core
Gaming / bettingchance, reward, tension, speeddice, gem, card, signal, trophy
Voice AIsound, rhythm, command, flowwaveform, mic, orb, speech path
Compliancetrust, order, rules, protectionseal, dog, badge, document, shield
Drone / roboticsflight, control, vision, missionwing, owl, crosshair, path, zone
Luxury / editorialtaste, material, ritual, restraintmonogram, seal, paper, emboss, mark
Productivityfocus, momentum, claritypath, check, block, calendar, light

If the project has a jtbd.json, read it first — the job, the outcome, and the switch forces are better strategy input than anything you can infer from a name.

Step 2 — Compose the board

Default output, unless the user says otherwise: one brand-kit overview image, a 3 × 3 grid, 4:3 or 16:10, clean presentation grid, consistent gutters, minimal text, every panel visibly connected.

Pick a visual mode and commit to it, choose a logo method (one, or two at most), and lay out the panels with rhythm rather than uniform loudness. See the Quick Reference above for all three.

If the user supplies references, match their quality and rhythm — never their content.

Anti-generic rules

Never produce: random floating icons · generic startup gradients · overdesigned logos · meaningless blobs · messy layout collages · fake tiny UI · inconsistent logo marks · too many colors · cheap neon · stock-template brand boards · corporate PowerPoint slides · soulless SaaS dashboards.

When a board is not working, the answer is almost never "more". Make it quieter, sharper, and more intentional. Before shipping, remove one accessory.

Step 3 — Explore directions, then hand off

  1. From the brief (or the project's jtbd.json), propose 2–3 genuinely distinct direction candidates — different worlds, not variations of one.
  2. Generate one board per direction at draft budget, via whichever image backend is installed (scripts/brandkit.py backends).
  3. The user picks a winner. Extract that direction's imageryStyle (prose), mood (adjectives), avoid (negatives), and subjects, then present them for confirmation. Never write without it.
  4. On confirm, run the handoff:
bash
# Existing token set: writes $extensions and regenerates DESIGN.md
python3 scripts/brandkit.py handoff --direction direction.json --tokens <set>.tokens.json

# No token set yet: saves brand-block.draft.json
python3 scripts/brandkit.py handoff --direction direction.json --out-dir <project-dir>

A saved draft is imported automatically by humane:design-tokens setup-edit run in the same directory — so the new token set is born with its art direction attached.

Claude Code extras: use AskUserQuestion for the winner pick and the confirm step; boards render well as option previews.

Tests

bash
cd brandkit && python3 -m unittest discover -s tests -v

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 Brandkit AI skill do?

This skill should be used when EXPLORING a brand identity that does not exist yet. Generates premium brand-guidelines boards for competing directions, and hands the WINNING direction off into the design-tokens brand block (confirm-then-write). Upstream of tokens; for producing assets under an EXISTING token contract use brand-illustrate instead. Carries reference direction for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Triggers on brand kit, brand board, brand directions, identity candidates, identity dire...

Why use Brandkit on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/glebis/humane-agentic-design/tree/main/humane/skills/brandkit. 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 Brandkit?

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 Brandkit?

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

Is the Brandkit AI skill free?

Yes. It is published on GitHub by glebis 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 👇