Information Architecture logo

Information Architecture

Community
mshadmanrahman
information-architecture

Map the user journey first, then derive the screen inventory, navigation and hierarchy from it, flagging any screen that serves no journey step. Writes the four-part markdown plus an HTML diagram of the step-to-screen mapping. Use after the brief is settled and before tokens or UI.

Overview

Publishermshadmanrahman
Repositoryclaude-for-designers
Skill nameinformation-architecture
Stars
40
Forks
10
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 mshadmanrahman on GitHub. Read the source before you install it.

Installation

Install the Information Architecture 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/mshadmanrahman/claude-for-designers/tree/main/.claude/skills/information-architecture
  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/mshadmanrahman/claude-for-designers.git /tmp/claude-for-designers
mkdir -p .claude/skills
cp -r /tmp/claude-for-designers/.claude/skills/information-architecture .claude/skills/information-architecture
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Information Architecture 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 Information Architecture 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 Information Architecture 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 structuring the product before any screens are drawn. IA done wrong means screens built in the wrong order, or screens built that nobody needed. Do not skip this.

Start by reading the design brief if one exists. If no brief is present, ask the user to run /design-brief first.

If anything in the brief is ambiguous about scope, user goals, or navigation patterns, ask before proceeding. It is better to clarify once than to structure the wrong thing.

Produce a structured IA document with four parts, in this order. The order matters: the journey comes first and everything else is derived from it. Do not begin with a list of screens.

Part 1: The journey

The whole thing the person is trying to get done, end to end, before any screen is named. Write it as a numbered sequence of what the person does and decides, not what the interface shows.

  • Name who is travelling. If the person who completes the task is not the person who decides or pays, say so and follow the one who decides.
  • Name where the journey actually starts. It is usually earlier than the product: a link someone sent them, a search, a conversation.
  • At each step, say what they are trying to find out or achieve.
  • Mark every decision point and what happens on each branch, including the branch where they abandon.
  • Name where it ends, and what "ended well" means for them.

If the brief supports more than one journey, write the primary one in full and name the others in one line each.

Part 2: Screen inventory, derived from the journey

Now list the screens. Every screen must name the journey step it serves. Use this shape:

Screen name: one line on its purpose (serves step N)

Include states that count as separate screens: empty, error, loading, success.

Then two checks, and report both explicitly:

  • Any screen that serves no step in Part 1. Say so plainly. It is probably not needed, or the journey is missing a step.
  • Any step in Part 1 with no screen. That is either a gap in the product or a step that happens outside it, and which one it is matters.

Part 3: Navigation pattern

How people move between the screens in Part 2. Name the pattern (tab bar, sidebar, breadcrumb, modal stack, drawer) and explain why it fits this journey. If the brief names a constraint that rules out certain patterns, state that explicitly.

Part 4: Content hierarchy per screen

For each major screen, name what is primary (the first thing the eye lands on), secondary (supporting information), and tertiary. Three levels maximum. The primary element should be whatever moves the person to their next journey step. Do not design the layout here; only the hierarchy.

Output: two files, not one

First, the four parts as a single markdown document, written to ia-map.md in the project folder. Use headers to separate them. That file is the deliverable and the source of truth, and /brief-to-tasks reads it next, so it has to stand on its own without the diagram beside it.

Then write the same mapping as an HTML page next to it, named ia-map.html. The reason is that the markdown buries the one thing this skill is for. The mapping is what matters: every screen names its step, and every step should have a screen. In markdown the two failures are two bullet lists plus a sentence, and the reader has to hold both lists in their head. On the page they are visible.

The page shows:

  • The journey as a numbered spine down the left, one box per step from Part 1.
  • Every screen from Part 2 hanging off the step it serves, joined to the spine by a line.
  • One screen serving two steps as one box with two lines into it.
  • A step with no screen: nothing attached to it. Give the gap its own colour and say in the box that no screen serves this step.
  • A screen that serves no step: a box with no line back to the spine, in a second colour.
  • A step that happens outside the product: attached, in a muted style, saying so. That one is not a gap.

Colour those two failures so the page shows them on its own, without a key telling the reader what to look for.

Do not write SVG by hand, and do not ask Claude to. Nobody can see the page while the code is being typed, so coordinates written blind collide in silence: text on top of text, a box past the edge, a line ending in empty space. It looks correct in the code and breaks when you open it. Build the page from plain HTML and CSS instead. A box is a div. The layout is CSS grid or flexbox. Every connecting line is a border on a cell that sits between the two things it joins, so the browser works out where the line goes and the line cannot miss. If a connector truly needs geometry, put a mermaid diagram inside the page rather than hand-written SVG.

Open the page in a browser before calling it done. Check four things: no box sits on top of another, no text escapes its box, every line touches both ends, and the page does not scroll sideways. Then check it again narrow, around 390px wide.

If the page and the markdown ever disagree, the markdown is right and the page gets fixed.

Frequently asked questions

What does the Information Architecture AI skill do?

Map the user journey first, then derive the screen inventory, navigation and hierarchy from it, flagging any screen that serves no journey step. Writes the four-part markdown plus an HTML diagram of the step-to-screen mapping. Use after the brief is settled and before tokens or UI.

Why use Information Architecture on TypingMind?

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

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/mshadmanrahman/claude-for-designers/tree/main/.claude/skills/information-architecture. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Information Architecture?

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 Information Architecture?

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

Is the Information Architecture AI skill free?

Yes. It is published on GitHub by mshadmanrahman 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 👇