Skip to main content
DesignKey Studio

Claude Code for Designers

· Part 1 of 3

Git, Claude Code, and why a designer should care

A designer-friendly mental model for Git, the free desktop app that lets you see every change without typing a command, and the shape of a good first prompt.

12 min read · Part 1 of 3

If you have ever opened a Figma file, changed a shadow on the hero card, and then wished you could see that change rendered in the actual product - this guide is for you. Claude Code turns that wish into a very short feedback loop. It takes plain-English instructions and applies them directly to the codebase so you can explore design decisions at the speed you think.

You do not need to learn how to use a terminal, and you absolutely do not need to memorize Git commands. You need a mental model, a visual tool to review changes, and enough prompt vocabulary to ask for what you want. That is Part 1.

The short version

A branch is a sandbox. Claude Code does the work inside the sandbox. GitHub Desktop shows you what Claude changed, file by file, before anything is shared. If a change looks wrong, you ignore it and ask again. Nothing on the real product moves until you and a developer both say yes. That is the whole safety model.

Why Git matters (and why you can mostly ignore it)

Git is a memory system for files. Every time you or Claude change something, Git takes a snapshot called a commit. You can rewind to any snapshot, start a new timeline from it, or merge two timelines together. Those parallel timelines are called branches. A pull request is how a branch gets reviewed and merged back into the main timeline.

The mental model of Git, minus the jargon
1

main

The live, trusted version of the project. Everyone sees it.

2

feature branch

A sandbox copy where Claude does the work. Affects no one but you.

3

commits

The snapshots Git takes of each change. Reviewable, undoable.

4

pull request

A formal "please merge this sandbox" - a developer reviews before it goes live.

You will hear all four words constantly. Internalize this row and you can follow any engineer conversation. You do not need to type a single command to work with them.

Here is the part that matters: when you work with Claude Code, you always work on a branch. If the output is wrong, you throw the branch away. Nothing on main changes until a pull request is merged. You cannot accidentally break the live product - the system literally does not let you.

Claude handles the branch for you

In practice you never type git anything. Claude Code creates the branch automatically at the start of a task, and commits as it goes. Your job is to describe the work and review the result. The Git mechanics stay invisible.

GitHub Desktop - your window into the changes

You do not have to rely on the terminal to see what Claude did. GitHub Desktop is a free app from GitHub that lists every changed file, shows you the exact lines that were added or removed, and lets you send the branch up for review with a button. It is the designer-friendly surface over everything Git does.

GitHub Desktop
Current repositorydesignkey-website1
Current branchfeature/hero-redesign2
Push origin25
3 changed files
  • components/Hero.tsxM
  • components/Hero.module.cssM
  • public/images/hero-bg.pngA
3
components/Hero.tsx+12 −4
24 const heading = useMemo(() => (    <h1 className="text-5xl font-bold">Join our community</h1>+    <h1 className="text-5xl font-bold">+      Welcome to the <RotatingWord words={words} />+    </h1>28  ), [words]);
4
Last fetched just now
GitHub Desktop after Claude finishes a session. The numbered areas below match the legend.
1

Your project

The repository you are working in.

2

Your branch

Your sandbox. Switch with one click.

3

Changed files

Every file Claude edited, listed here.

4

The exact lines

Red is removed, green is new. Read before approving.

5

Push

Sends your branch to GitHub for review.

Three things to know when you open GitHub Desktop for the first time:

  1. 1

    The left pane is your review queue

    Every file Claude edited appears here. Click one to see the diff on the right. Uncheck a file to exclude it from the commit - a fast way to cherry-pick what you like.

  2. 2

    The right pane is the source of truth

    Red lines were removed, green lines were added. You do not need to understand the syntax - scan for obvious mistakes, hardcoded text that looks wrong, or missing alt text on images.

  3. 3

    The commit box is your message to the team

    Claude usually writes this for you. Read it. If the message says something different from what you asked, the change may not match your intent.

GitHub Desktop is optional, not required

For most sessions you will review changes inside Claude Code itself - it shows every edit before writing. GitHub Desktop is handy when you want a clean visual summary or you have come back to a project after stepping away.

Or skip the app entirely and ask Claude

Everything GitHub Desktop does, you can also ask Claude Code to do. These are not technical commands you need to memorize - they are sentences you say in a chat. A short starter set:

You sayCreate a new branch for this work and switch to it.

Claude doesClaude makes a branch with a sensible name so your changes stay isolated from main.

You sayPull the latest changes from main into my branch.

Claude doesClaude fetches, rebases or merges, and tells you if anything conflicts.

You sayShow me what has changed on this branch so far.

Claude doesClaude lists every changed file and summarizes what was edited in plain English.

You sayCommit everything and push this branch.

Claude doesClaude stages the files, writes a sensible commit message, and pushes.

You sayOpen a pull request with a good title and test plan.

Claude doesClaude drafts the PR body, writes the test plan, and creates the PR for you.

You do not need to learn Git

Nothing above assumes you know a single git command. If you forget the exact phrasing, ask “how do I put my changes up for review?” and Claude will tell you and do it. The chat interface is the interface.

What Claude Code actually is

Claude Code is Anthropic’s coding assistant. You open it inside a project, describe what you want in plain English, and Claude reads the relevant files and edits them. It pauses before every change so you can approve or reject it.

Think of it as a very patient junior developer who has already read the entire codebase, will never miss a typo, and asks permission before doing anything risky.

The three tools you will have open

Claude Code

Where the conversation happens

Runs inside your editor. You type a request, it writes the code.

GitHub Desktop

Where you see every change

Free, visual, and designed for people who do not want to type commands.

Browser

Where you verify the result

Chrome with the preview URL open. Claude can even drive Chrome for you.

Your first session, in designer terms

Forget about commands. The actual rhythm of working with Claude Code is five steps, and only the middle three involve you at all:

The five-step rhythm of a Claude Code session
1

Claude creates a safe sandbox

A new branch is made automatically. Nothing you do here can affect the live product.

2

You describe what you want

Plain English. "Rework the hero to use a rotating word, keep everything else the same." Files, outcomes, constraints.

3

Claude previews every change

You see a diff for each file before it is written. Approve the good ones, reject the wrong ones, redirect when needed.

4

You review the final result in the browser

Claude tells you how to see the live preview (usually http://localhost:3000 or similar). Click around, check edge cases.

5

A developer opens the pull request

Click "push" in GitHub Desktop to send the branch up. A developer does a final review and merges. Done.

Steps 1, 3, and 4 are automatic or visual - no typing required. Only steps 2 and 5 need attention from you.

You will be asked for permission a lot - that is the point

Claude Code pauses before every file write, every shell command, and every network call. The pause is not a bug; it is the safety net that lets you catch a wrong turn before it becomes 500 lines of wrong code. Read the preview, then approve or reject. Do not hammer “yes.”

What a good prompt looks like

You do not need to know the codebase, and you definitely do not need to know file names. Designers describe changes the way designers already think — point at a page, point at a section on that page, and say what should change. Claude finds the files on its own.

Claude Code - session

> On /about, in the hero section, replace the static headline with a rotating word carousel. Keep the typography and colors exactly as they are.

I’ll find the component that renders the hero on /about, read it, and preview the edit before writing.

[Claude locates the file, reads it, shows the planned change]

Where

Page URL + section name

"On /about, in the hero section" is the whole locator Claude needs. No file paths.

What should change

Describe the visual outcome

What should a user see after the change? Be specific about the visual, not the code.

What should not

Protect what you like

"Keep the typography and colors unchanged" prevents Claude from drifting into a redesign.

Copy-ready prompts are on the cheat sheet

We have a companion page with seven copy-ready prompts for the most common design tasks - iterating a component, generating variations, matching a Figma frame, auditing for accessibility. Bookmark it and reach for it the first few sessions. Open the cheat sheet →

Things worth noticing (not rules to follow)

We want you working fast. In practice that means running Claude Code in auto mode — it stops asking you to approve every tiny step and just does the work, the way any thoughtful collaborator would. You review the final result, not the individual edits.

The cheat sheet has the exact command to launch in that mode. For now, here are the few things it is still worth paying attention to:

  1. 1

    Check the outcome, not the code

    Open the preview in the browser. Walk through the flow you asked about. If it looks right on desktop and mobile, you are done. You do not need to read the code.

  2. 2

    If something looks off, just ask Claude to revert

    Nothing is permanent until the branch is merged. "Undo the last change" or "go back to how it was before" gets you back to a clean slate in seconds.

  3. 3

    Let a developer merge the PR

    Opening a pull request is yours. Clicking "merge" is a developer's job - not a gatekeeping ritual, just a second pair of eyes before the change goes live.

  4. 4

    If Claude wants to do something dramatic, pause

    If it suggests resetting the project, deleting files, or force-pushing, that is the moment to ask in the design or engineering channel before proceeding.

Auto mode is on the cheat sheet

The launch command that skips per-step permission prompts, plus the command to resume a previous session, live on the cheat sheet. Use auto mode from day one; you can always stop a session with Ctrl+C if something looks wrong.

What this actually produces

Enough theory. The screenshots below are all from a single Claude Code session where a designer asked for variations of a countdown card on the Tampa JUG community site. The full case study is Part 2, but here is a preview of what one session can produce in under an hour:

V1 Luma - clean white event card with a square poster on the left and stacked info on the right.
V1 - a Lu.ma-inspired clean card. First variant generated from a single plain-English prompt.
V7 Split Gradient - brand radial-gradient left panel with giant date, white info panel right.
V7 - signature brand radial gradient with an oversized date. Four prompts away from V1.
V9 Radial Hero - signature brand radial gradient background with the event poster contained on the right.
V9 - a more hero-scale take on the same signature gradient.
V12 Neon - deep navy card with glowing brand-blue countdown digits and a poster frame.
V12 - the same event, re-skinned in a dramatic neon treatment on brand.

These are not mockups. They are production React components.

Each variant is a real .tsx file in the project. They render with live event data pulled from Meetup’s API. Swapping the home page from one variant to another is a one-line change. That is the leverage Claude Code gives you.

Where AI gets it wrong (and why that is fine)

Honesty check. In the same session that produced the screenshots above, Claude also made three mistakes worth calling out. None of them reached production, but all three are instructive.

Mistake 1 - Wrong aspect ratio

The poster got cropped on five variants

Claude assumed the Meetup event poster was 16:9. It is actually square-ish, so the top and bottom were sliced off. The fix took one prompt: "replace the cropped fit with a contained fit and add a blurred backdrop to fill the empty bars." Caught in review, not in production.

Mistake 2 - Off-brand palette

The Magazine variant used a cream background

Claude interpreted "editorial magazine aesthetic" too literally and reached for a print-warm cream. On inspection, it was off-brand. One sentence redirected it to the brand neutral. Again, caught in review.

Mistake 3 - Overconfident self-review

It thought the first five variants were good

Claude self-evaluated the first pass as polished. A human spot-check disagreed - the cropping, the weak hierarchy on two of the five, all needed rework. The lesson: every Claude-generated batch needs a human review pass, no matter how confident the model sounds.

The rule is the same rule you already use

A junior designer would make these mistakes. A junior developer would make these mistakes. Claude makes these mistakes. The job of a senior contributor is to catch them in review. Claude just makes the cost of catching them much, much lower, because the turnaround is seconds instead of days.

The cheat sheet lives on its own page

Everything you need as a quick reference - the nine slash commands worth remembering, copy-ready prompt patterns for seven common design tasks, and links to the UI/UX skills on GitHub we recommend bookmarking - is on a dedicated page so you can keep it open in a tab.

Next up

Part 2 - Generating design variations (the TJUG case study)

One countdown card, fourteen variations, real feedback loops. Watch the same component redesigned end to end, including the moments Claude got it wrong.