hello, Janika_

Welcome to
the world of
Claude Code.

A gentle guide to building things you’ve only imagined, starting with that tube widget you’ve always wanted.

Part 1 of 3, made for 
→ edit the name if you like, press enter to begin
A note before we begin, Janika

You don’t need to know how to code to build something real.

You already know how to brief a team, hold a line on a drawing, and spot a badly detailed junction at twenty paces. That instinct is the job here. Claude does the typing.

Think of Claude Code as a brilliant graduate architect on your team who types at ten thousand words a minute and has read every building manual ever written. They need a brief. They need rules. They need you to check the drawings. Give them those three things and they’ll build anything.

By the end of this guide you’ll have built your first real thing. Take your time. Every pause point in this deck is a chance to try it yourself.

A leather notebook open on a desk with a fountain pen and a dried lavender sprig
What we’ll build together

A tube
widget.
Your widget.

Live delay times. Next arrivals. The lines you actually take, in one glance on your lock screen.

We’ll start with a simple web version you can play with today, then turn it into something that lives on your iPhone.

Your Lines LIVE
Central 2 min
Circle 5 min
Victoria 8 min
Line Status
District Good
Metropolitan Minor delays
Chapter 01 · The basics

So what is Claude Code?

Imagine a coding partner who:

  • →  Lives in your terminal (that black window with text)
  • →  Reads and writes files on your computer
  • →  Runs commands for you
  • →  Never gets tired or annoyed
  • →  Works at whatever pace you do

That’s Claude Code. A chat window, connected to your computer, with Claude on the other end.

~ claude
> can you make the button bigger?
… thinking
Done. I updated the padding from 12px to 20px.
> and change the colour to red
Changed to #E32017. Refresh to see it.
Chapter 02 · How it works

The whole thing,
in one picture.

Three players. That’s it.

You
the architect
Claude
the contractor
Your Mac
the site

You brief Claude. Claude builds on your computer: reading files, writing code, running commands. You watch, comment, and steer. Same dynamic as you and a good contractor, just compressed into minutes instead of months.

Pause · Hands-on #1

Let’s install it.

This takes about 5 minutes. Tick each step off as you go.

Setup · One-time only

Install your tools.

We’re installing two things. Claude Code is the brain. Superset is a beautiful Mac app that lets you run Claude (and friends) in a calm, organised way, one project per window, without touching Terminal much at all.

0. Make an Anthropic account Open claude.ai in your browser and sign up. You’ll use this to log in. Claude Code charges a few pennies per message, no surprise bills for a small project like this.
1. Install Apple’s developer tools (one-time Mac setup) Apple ships Macs without Git pre-installed. Before anything else, open the Terminal app (press ⌘ + Space, type “Terminal”, Enter). Paste this and hit return:
xcode-select --install
A macOS dialog will pop up asking to install Command Line Tools. Click Install, accept the licence, and wait 5 to 10 minutes. It’s about 2 GB. When it finishes, everything else will just work. (If it says “already installed”, you’re set, move on.)
2. Install Claude Code Still in Terminal, paste this one line and hit return:
curl -fsSL https://claude.ai/install.sh | bash
(Anthropic’s official installer. Handles the lot. Takes under a minute.)
3. Download Superset In your browser, open superset.sh. Click the big Download for macOS button. Open the file, drag Superset into your Applications folder, then launch it.
4. Make a workspace for today’s project Inside Superset, click New Project (or New Workspace). Name it tfl-widget.
Change the location first Before you click Create, change the Location. The default is ~/.superset/projects, which is hidden on macOS (anything starting with a dot is invisible in Finder). You won’t be able to find your project files later. Pick somewhere you can see, like ~/Desktop or ~/Projects. Type the path in, or click the folder icon next to the Location field.
Superset creates a tidy folder there, plus its own little work area (called a git worktree; think of it as a sealed-off job site so nothing spills into other projects).
Already made one in the hidden spot? No harm done. Open Finder, press ⌘+Shift+G, paste ~/.superset/projects/tfl-widget, hit Enter. Your files are all there. For future projects, use the Desktop path above.
5. Pick Claude as your agent Superset asks which agent you want to run. Choose Claude Code. First time only, it’ll ask you to log in with your Anthropic account.
If Superset drops you into a plain terminal The Claude preset sometimes doesn’t connect the first time. No panic. Paste this one line to make your own shortcut called opus:
echo 'alias opus="claude"' >> ~/.zshrc && source ~/.zshrc
Then type opus and hit Enter whenever you want to start Claude. If it says “command not found”, use the explicit path:
echo 'alias opus="$HOME/.local/bin/claude"' >> ~/.zshrc && source ~/.zshrc
6. First-run setup (2 minutes, one-time only) The first time Claude starts, it walks you through a couple of quick screens. Use arrow keys and Enter to navigate.
Screen 1 · Pick a theme Dark, light, or auto. Auto matches your Mac’s system setting. Pick whichever reads best. You can change it later with /theme. Screen 2 · Log in Claude prints a URL and says “open this in your browser”. Do that. In the browser, sign in to your Anthropic account (the one from Step 0), approve the connection, and it’ll tell you to return to Terminal. Come back, press Enter, you’re in. Screen 3 · Trust this folder Claude will ask if it can read and edit files in the current folder. Say yes. (Only ever do this for folders you own and trust.)
7. Say hello You’re now in the Claude chat. Type:
hi, are you ready to build something lovely?
Why Superset?

Same Claude, but with a proper home. You can see your files, the changes Claude makes, and the chat, all in one window. Later you can even run several projects at once in separate tabs. Think of Superset as the site office: Terminal was the open pavement.

Stuck on a step? Take a breath. Show whoever’s guiding you the exact error you see, they can help.

Pause · Hands-on #2

Say hello.

You’re in Claude Code. A blinking cursor is waiting. Let’s warm up.

Try these · One at a time

Your first three prompts

Type each one, press Enter, watch what Claude does. No wrong answers, you’re just getting a feel.

1. Introduce yourself
hi, who are you and what can you help me with?
2. Ask it to do something simple
create a file called hello.txt with a nice welcome note inside
Then check your folder, the file is really there.
3. Ask it to change the note
make the welcome note more cheerful and add three emojis
What just happened

Claude understood what you wanted, wrote to your computer, and told you what it did. That’s the whole loop. Everything else is just scaling this up.

Chapter 03 · The words

Six words
to start with.

Learn these and you’ll understand most of what people say about AI coding. Don’t memorise them, we’ll use each one as we go, and a few more will show up along the way.

Context01
Everything Claude can see in this exact conversation. The chat, the files you’ve shown it, the errors it’s seen. Like a whiteboard you share, wiped clean when you start again.
Memory02
Notes Claude keeps for itself, across sessions. Things you’ve asked it to remember, your preferences, your project rules. Like a notebook that survives the day ending.
Prompt03
Whatever you type to Claude. A question, an instruction, or a ramble. The better you explain, the better Claude performs.
Model04
The specific brain you’re talking to. Claude has a few, some are slower and deeper, some are fast and sharp. Like picking between a novelist and a sprinter for the job.
Tool05
A specific thing Claude is allowed to do. Read a file. Write a file. Run a command. Search the web. Each one is a verb Claude can use on your behalf.
Agent06
A helper Claude can spin up to handle a specialist task, a reviewer, a researcher, while the main Claude keeps the plot. Like delegating to a teammate for a focused job.
Chapter 04 · Claude’s short-term memory

Context is
the whiteboard.

Every conversation has a shared whiteboard. Claude can only see what’s written on it right now.

What goes on the whiteboard: your messages, the files you’ve shown it, the outputs of commands it’s run.

What does not: yesterday’s chat, other projects, anything on the internet it hasn’t been shown.

Claude
sees this
Rest of the world
The whiteboard
Why it matters

The whiteboard fills up. When it does, Claude starts forgetting the earliest things you said. Keep chats focused. Start a fresh session for each new topic.

Chapter 05 · Claude’s long-term memory
~ CLAUDE.md
# TFL Widget Project

## About me
I live near Oxford Circus.
I take the Central line mornings
and Victoria line evenings.

## Rules for Claude
- Keep the design clean and modern.
- Use the real TFL line colours.
- Ask before installing anything new.

## Style
- Rounded corners, soft shadows.
- San Francisco font (the Apple one).

Memory
is a note
to itself.

The whiteboard gets wiped. Memory doesn’t.

There’s one magic file called CLAUDE.md. Whatever’s in there, Claude reads at the start of every session.

Tell it who you are, what you’re building, and how you like things. It’ll remember next time too.

Chapter 06 · The one file that matters most

CLAUDE.md is
your project code.

Every construction project has a code sheet: the rules, the conventions, the “how we do things here”. CLAUDE.md is the same thing for your project, and Claude reads it every single time you start a session.

Get this file right and Claude feels like it’s known the project for months. Get it wrong, or bloat it, and every session starts with it noisily clearing its throat.

CLAUDE.md 60 lines About the project Rules for Claude Style & conventions A1

The ideal CLAUDE.md, one page, tidy

Five rules straight from the Claude Code team.

Boris Cherny (the engineer who built Claude Code) and his team share CLAUDE.md across the whole Anthropic codebase. Here’s how they keep it healthy.

01

Keep it short.

Aim for under 300 lines. The best ones sit under 60. Every line goes into every session, so every line is paying rent.

02

Only universal things.

Don’t put rules that only apply sometimes. If it only matters in one corner of the project, make a second CLAUDE.md inside that folder.

03

Grow it by correction.

When Claude makes a mistake, end your correction with: “Update your CLAUDE.md so you don’t make that mistake again.” It is strangely good at writing rules for itself.

04

Let /init start it.

Type /init in Claude and it writes you a sensible first draft based on what your project looks like. You refine from there.

05

Prune it monthly.

Ask Claude: “read our CLAUDE.md and suggest what we can cut. We want it tighter.” A fresh pair of eyes keeps the project code readable.

06

Commit it to Git.

CLAUDE.md belongs with the project, not on your Mac. When you save checkpoints, it gets saved too. Future-you will thank past-you.

Architectural parallel

Think of CLAUDE.md as the project execution plan: site constraints, office standards, approved materials, responsibilities. Read on day one of every stage. Grows only when a lesson learned earns its place.

Chapter 07 · The five lines that matter most

Five rules
to start with.

Copy these into your project’s CLAUDE.md on day one. Delete anything that doesn’t apply to you. Grow the rest as you go.

## TIER 1 : Five rules for day one ### 1. The spec wins every conflict. If a plan or piece of code contradicts the spec, the spec wins. Stop and flag it. ### 2. Explain before you edit. Before changing any file, tell me what you plan to change and why. No surprise diffs. ### 3. No stubs, no placeholders. If you can’t implement something fully, stop and ask. No “TODO” markers. ### 4. Stay inside the current scope. Build only what the current phase specifies. Note discoveries elsewhere. ### 5. Explain before you change course. If the planned approach is blocked, stop. Explain the problem, present 2-3 options, wait for a call.

→ Full 14-rule starter: templates/CLAUDE.md.starter in ship-it. Tests-before-code and nine other rules live there, ready when you are.

Why CLAUDE.md exists at all

There’s a well-known pattern in AI-assisted coding called the 70% problem: AI will happily get you to something that almost works, fast. The last 30% (edge cases, consistency, the little details) is where projects stall. CLAUDE.md is how you close that gap. Every rule you put in it is a small piece of taste Claude can now share with you.

Chapter 08 · How Claude thinks about size

Tokens, context,
and the bill.

Everything Claude does costs a tiny amount. Before we go any further it’s worth ten minutes on how that works, so nothing surprises you.

A token is about four characters.

“Build me a tube widget” Build me a tube widget 5 WORDS = 5 TOKENS, ROUGHLY. 23 CHARACTERS. ABOUT 4 CHARS PER TOKEN.

Short message: a few tokens. A paragraph: about 100. A whole code file: several thousand. Claude’s “whiteboard” (the context window) currently holds up to a million tokens. That sounds like a lot until you start loading a whole codebase into it.

Target 50% context. Past 80%, Claude starts compacting.

When the whiteboard is half-full, Claude is at its sharpest. Past 80%, it starts compacting the earliest parts of your chat to make room, summarising what came before rather than carrying every word. Keep chats focused. Start a fresh session when you switch topics.

Costs: two ways to pay.

01

Pay-as-you-go (API)

Every message costs a fraction of a penny. No caps. You top up credit as needed. Great for low usage or weekend tinkering.

02

Subscription (Claude Code Max)

A flat monthly price with a generous daily message limit plus a weekly one. Better if you’re coding every day. If you hit the cap, Claude pauses you until the window resets.

In practice

A well-written CLAUDE.md saves you tokens because Claude starts every session oriented instead of having to re-read half the project to catch up. Writing it well is a cost control, not a cost.

Check where you’re at: /usage

Any time you want to know how much of your plan’s daily or weekly allowance you’ve used, just type /usage in the chat and hit Enter. Claude shows you:

  • →  Your current plan (Free, Pro, Max, Team)
  • →  How much of the daily limit is used, and when it resets
  • →  How much of the weekly limit is used, and when it resets
  • →  This session’s spend so far

Peek at it once a day while you’re getting used to the rhythm. After a week or two you’ll have a feel for what’s expensive and what isn’t.

If you hit the limit

Claude tells you when the window resets (usually a few hours). You can either wait, or upgrade your plan at claude.ai/settings/billing.

The status line: your dashboard.

When Claude Code is running in your terminal, a little strip at the bottom of the chat shows you everything at a glance.

Opus 4.7 (1M context) on main [24% used] ▸▸ auto mode on (shift+tab to cycle) THE MODEL + ITS WHITEBOARD SIZE YOUR GIT BRANCH CONTEXT USED

To turn it on, you don’t edit any files yourself. Just tell Claude what you want to see. In the chat, type:

/statusline show the model, git branch, and context used so far

Claude Code writes a tiny script, saves it into your settings, and the status line appears at the bottom of the chat after your next message. No JSON to paste, no file to create.

Hands-on

In your tfl-widget folder, run the command above. Wait a beat, then look at the bottom of the chat. Point at each segment in turn and say aloud what it tells you. That’s the dashboard you’ll glance at a hundred times a day.

Chapter 09 · Shift + Tab

Four modes,
one shortcut.

Every time Claude wants to edit a file or run a command, it pauses and asks. That’s the default, and it’s a good place to start. Sometimes you want fewer pauses, sometimes you want none. Four modes give you that dial.

Press Shift + Tab in the chat to cycle. The status line at the bottom tells you where you are.

Opus 4.7 (1M context) on main [24% used] > _ DEFAULT Opus 4.7 (1M context) on main [24% used] ▸▸ accept edits on (shift+tab to cycle) ACCEPT EDITS Opus 4.7 (1M context) on main [24% used] ▸▸ plan mode on (shift+tab to cycle) PLAN Opus 4.7 (1M context) on main [24% used] ▸▸ auto mode on (shift+tab to cycle) AUTO

The bottom of your chat, one row per mode. Shift+Tab moves you down the list and back up.

01

Default

Claude asks before every edit and every command. Best while you’re learning, or any time you’re touching something important. The status line stays quiet.

02

Accept edits

Claude edits files in your project without asking. It still asks before running commands. Best when you trust the direction and would rather review the diff after. Status line shows ▸▸ accept edits on.

03

Plan mode

Claude reads, researches, and writes a plan, but doesn’t touch anything. When the plan looks right you approve it, and it starts building. Best before a big change. Status line shows plan mode on.

04

Auto mode

Claude runs without asking. A background safety check blocks the risky things: production deploys, pushing to main, deleting files it didn’t create, running code pulled from the internet. Best for long, well-scoped tasks you trust. Needs a Max, Team, or Enterprise plan. Status line shows ▸▸ auto mode on.

Boundaries still work

Even in auto mode, if you tell Claude “don’t push until I say so” or “wait for my review before deploying”, it honours that. The safety check treats those lines as hard stops until you lift them in a later message.

Hands-on

Press Shift + Tab a few times in your chat. Watch the status line at the bottom change as you cycle. Land back on default when you’re done exploring, that’s the mode you’ll want for most of Part 2.

Chapter 10 · A small helpful detail

The cache
(the one other word).

Claude has a kind of muscle memory. The first time it reads something, it’s slow. Reading the same thing again right after? Fast and cheap.

That’s the cache. You don’t need to manage it. Just know: staying inside one focused conversation is faster than restarting over and over.

In practice

Finish a thought before switching topics. Your wallet and your patience will thank you.

Chapter 11 · The three Claudes

Three brains.
Pick one for the job.

There isn’t just one Claude. There’s a family. Each has its own speed, depth, and cost. Most of the time you won’t need to choose, Claude Code defaults well. But it helps to know.

O
Opus 4.7

The architect.

Deep thought, slow steps.
Use for Hard thinking. Big decisions. Planning a whole new feature. Gnarly bugs.
S
Sonnet 4.6

The hero.

Balanced, quick and smart.
Use for Day-to-day building. The default for most conversations. Ships work fast.
H
Haiku 4.5

The sprinter.

Very fast, very focused.
Use for Tiny tasks. Quick lookups. When you want an answer in seconds.
Rule of thumb

Stick with Sonnet for most of today. Reach for Opus when something feels hard and you want real thought. You can switch any time by typing /model, the / at the start tells Claude it’s a built-in shortcut, not a message.

(Version numbers move quickly. Check claude.ai for the current family if you want the freshest one.)

Chapter 12 · What Claude can actually do

Tools are the verbs.

Claude talks to you in English, but inside, it works through a small set of tools. Here’s the family.

01

Read

Open a file and look at it. The most basic thing, Claude can see what’s already there.

02

Write

Create a brand new file. Claude fills it in with whatever you’ve asked for.

03

Edit

Change a small part of an existing file. Faster and safer than rewriting the whole thing.

04

Run

Run any command in the terminal, installing things, starting a server, peeking at what’s in a folder. (Inside Claude this is called “Bash”, same thing.)

05

Web Search & Fetch

Look things up on the internet. Read a specific page. Fetch a live API, like the TFL one.

06

Task (Agents)

Send a sub-job to a helper. More on this in a moment.

Permission

The first time Claude uses a tool in a new folder, it asks “is this okay?”. That’s your safety net. Say yes if you trust it, no if you want to check.

Chapter 13 · Delegation

Sub-agents are
specialists.

When a job is big or distracting, Claude can spin up a helper with a single purpose, research this, review that, and get a clean answer back.

Why it matters: your main Claude stays focused on the thread of work. The helper handles the side quest.

Main Claude
the conductor
Researcher
reads docs
Reviewer
checks work
Explorer
finds files
Chapter 14 · Shortcuts with superpowers

Skills are
standard details.

An architect doesn’t re-draw the same parapet flashing on every job. You keep a library of standard details and call them by name.

A skill is the Claude version of that: a named, reusable workflow you invoke by typing a forward slash. Instead of typing out the same instructions every time, you type one word.

~ claude
> /init
(drafting your CLAUDE.md)
Done, 48 lines, saved.
> /commit
(reading diffs, crafting message)
Committed: “add project code to CLAUDE.md”
> /review
(examining your branch)
Looks good. One suggestion, line 34.

Three things to know.

01

Type a slash.

Typing / in the chat brings up a list of available skills. You don’t have to memorise them.

02

Built-in, then your own.

Claude ships with a handful (/init, /model, /clear). You’ll learn to add your own in Part 2.

03

They’re just files.

Each skill is a tiny file in your project that tells Claude what steps to run. Like a standard detail in a details library.

Rule of thumb

If you find yourself typing the same instructions twice, that’s a skill waiting to happen. Part 2 will show you how to make them.

Chapter 15 · How to talk to Claude

Prompting is
just good briefing.

Imagine briefing a smart friend who’s brilliant but has never met you. What would they need to know?

Weak

“Make me a TFL widget.”

Too vague. Claude will guess at everything, which lines, what style, what data, and you’ll spend the next hour undoing its guesses.

Strong

“Build a web page that shows next arrivals for the Central and Victoria lines at Oxford Circus. Use real TFL colours. Keep it minimal. Refresh every 60 seconds.”

Specific. Scoped. Claude knows what to build and what to leave out.

The three things a good prompt has

1. What, the thing you want
2. For whom / where, context Claude couldn’t guess
3. Constraints, colours, tone, size, tech, anything you care about

Try this in your terminal I’m going to build a TFL widget. Before we write any code, ask me five questions that will help you understand exactly what I want.
Chapter 16 · The one catch
!

Sometimes
Claude makes
things up.

It’s called a hallucination. Claude will confidently say a function exists when it doesn’t, or invent a library that was never built.

It’s not lying. It’s pattern-matching, and sometimes the pattern is plausible but wrong.

01

How to spot it

It sounds perfectly reasonable but something feels off. A library name you’ve never heard of. A command that fails in a weird way. Too-clean-to-be-true answers.

02

How to prevent it

Ask Claude to show its sources. Tell it to read the real docs first. Say “if you’re not sure, say so.” A good prompt beats a hallucination.

Try this Before we use any library, please check it actually exists by fetching its real documentation. If you can’t verify it, tell me so.
Chapter 18 · Saving your work, properly

Git is your
revision stamp.

Every architect knows the fear of a “final_final_v7_Janika_use_this_one.dwg” moment. Git stops that happening. It gives every change a revision, a reason, and a way back.

01

Git · the revision log

A tool that lives on your Mac. Every time you save a checkpoint, Git stamps it with who, when, and what changed. You can rewind to any stamp, any time.

02

GitHub · the office server

A website that stores your Git history in the cloud. A safe copy that survives your laptop dying, and lets others see your work.

The four words you’ll hear.

Commita saved revision
A checkpoint. You made some changes, they work, you stamp them with a little message (“add Central line arrivals”). Now they’re in the revision log forever. Like issuing a drawing: dated, described, filed.
Pushsend to the cloud
Once you have commits on your Mac, pushing uploads them to GitHub. This is your offsite backup and how the world sees your work. Like sending issued drawings to the office server.
Brancha parallel draft
A side-copy of the project where you try something without affecting the main version. If it works, you merge it in. If it doesn’t, you throw it away. Like a sketch study you do on tracing paper over the working drawing.
Pull Request“please merge my branch”
A formal ask on GitHub: “here are my changes, please review them, then fold them into the main version.” Reviewers can comment, you can tweak, then everyone agrees before the merge happens. Exactly like a drawing issue for comment: RFIs get raised, you respond, then it’s approved for construction.

The whole lifecycle, in one picture.

Edit YOUR MAC Commit LOCAL LOG Push GITHUB Pull Request REVIEWED Merge LIVE

Worktrees: two hands on the pencil.

A worktree is a second copy of your project on your Mac, linked to a different branch. Two Claudes can run in parallel on two different features without stepping on each other. Like giving two junior architects separate drawing sets for separate rooms so neither blocks the other.

MAIN merged feature-a Worktree A, Claude 1 working here feature-b Worktree B, Claude 2 working here
The good news

You don’t need to learn any Git commands. Just say to Claude: “please commit this”, “push to GitHub”, “open a pull request”, or “make me a worktree for this” and it handles the ceremony. You stay at the level of intent.

Chapter 19 · Now, the fun part

Let’s build
your widget.

We’ll go in four small steps. Each one ends in something you can see and play with.

01

A written plan

Ten minutes with Claude to agree on: which lines, which stations, what it looks like, what it needs to do. This is your spec.

02

A working web version

A simple web page that shows live arrivals from TFL’s real API. Runs in your browser in one click. We’ll see it tonight.

03

Polish it

Real colours. Nice typography. Loading states. The little touches that make it feel like a product.

04

The iPhone widget

Turn the web version into a real widget on your lock screen. This is its own adventure, a project for another evening.

Today’s target

Steps 1 and 2. One hour. Something you can show your friends tonight. That’s a win.

Before you build · TFL API key

TFL’s live tube data is free, but they want you to register for an app key (a small password so they know who’s asking). Five minutes, with a couple of small traps. Follow these in order.

1 · Register Open api.tfl.gov.uk in your browser. Click Register in the top right. Fill in name, email, password. 2 · Check your spam folder TFL sends an activation email. Nine times out of ten it lands in spam or junk. Go to that folder, find the email, click the Confirm your email link inside. Your account is now active. 3 · Log in Back on api.tfl.gov.uk, click Sign In. Use the account you just activated. 4 · Go to Products Click Products in the top navigation. You’ll see a small list. You want the one called 500 Requests per min. (Ignore TrackernetFeedPublic, it’s a different service.) 5 · Subscribe, get your key Click into 500 Requests per min. You’ll see a subscribe form. Give it any name, tfl-widget is fine, then press Subscribe. The page now shows your Primary key and Secondary key. Either one works. 6 · Keep it somewhere handy Copy the Primary key into your notes app or a stickies. Claude will ask for it when you kick off the build on the next slide. Treat it like a password, don’t paste it into a public place.
Deep breath

You can’t break your Mac doing this. The worst that happens is some code doesn’t work, and we start that bit over. Messy is normal. Nothing is final.

Pause · Hands-on #3

Kick it off.

You’re in the tfl-widget folder with Claude running. Time to steer.

Step 1 · The plan

Start with a conversation, not code.

Copy this and paste it to Claude. Then chat back and forth until you both agree on the plan. These prompts are starting points, change the words to suit you.

Paste this first I want to build a small, beautiful TFL tube widget. It should show me the next arrivals for my regular lines, and the live status of each line (good service, delays, etc).

I’m not a coder. Talk to me like a friend who’s about to design this with me.

Before we write a single line of code, ask me five questions that will shape the design. Things like which lines I care about, which stations, how it should look, how often to refresh. Keep it simple.
Step 2 · Lock it in

Save the plan to memory.

Once you’re happy, ask Claude to write it down so it doesn’t forget:

Paste this next Great. Please write everything we just agreed into a file called CLAUDE.md at the top of this folder. That way every future session starts with the plan already in mind.
Step 3 · Build the first slice

Ask for the smallest working thing.

Paste this last Let’s build the smallest possible version first. Just a single web page called index.html that fetches live Central line arrivals from the real TFL API and shows them as a simple list.

Here is my TFL app key: <paste your key here>

No styling yet, plain and ugly is fine. Once it works, we’ll make it beautiful. When you’re done, tell me exactly how to open it in my browser.
Step 4 · See it live

Open your creation.

Once Claude says it’s done, find the index.html file in your tfl-widget folder (use Finder) and double-click it. Your default browser will open and, hopefully, show live tube times.

Didn’t work? Not a problem. Paste the error or the blank page back to Claude and say: “Here’s what I see. What do you think is going wrong?”

If Claude gets stuck

Say: “Show me the error in your own words. What do you think is going wrong?”

Appendix · Keep this open

The little
dictionary.

Every word you’ll hear, in plain English. Bookmark this page. You’ll come back to it.

Terminal
The black window with text where you type commands. Claude Code lives here.
Command
A line of text you type to make the computer do something. Like “open this folder” but written in computer language.
Folder (Directory)
Same thing. “Directory” is the old word. Your projects live in folders.
File
A single document. Code is written into files. You’ll have lots of them.
Context
Everything Claude can see right now in this one conversation. The shared whiteboard.
Context window
The maximum size of the whiteboard. When it fills up, the oldest things fall off.
Memory
Notes Claude keeps between sessions. Usually stored in a file called CLAUDE.md.
CLAUDE.md
A special text file where you write rules, preferences, and project info. Claude reads it automatically every time.
Prompt
Whatever you type to Claude. Could be a question, an instruction, or just a thought.
Model
The specific version of Claude you’re talking to. Opus, Sonnet, Haiku, different speeds and depths.
Tool
A specific action Claude can take, reading a file, running a command, fetching a web page.
Agent / Sub-agent
A helper Claude spawns to handle a focused side task, like research or reviewing.
Hallucination
When Claude confidently says something that isn’t true. Rare, but worth watching for.
Cache
Claude’s short-term speed-up. Recently-read things are remembered briefly so they’re faster next time.
API
A way for apps to talk to each other. TFL has an API that gives out live tube data.
Git
A tool that gives your project folder perfect memory. Save versions of your work so you can always rewind.
Commit
A saved checkpoint. Like saving your place in a game. You can always come back to this exact version.
Repo (Repository)
A project folder tracked by Git. Every version you’ve ever saved is remembered.
Skill / Slash command
A shortcut you type like /review to run a pre-made mini-workflow.
Hook
An automatic check that runs at a certain moment. Like a safety rule: “warn me before deleting anything.”
MCP Server
A plug-in that gives Claude new abilities, talking to your calendar, browsing websites, etc. Advanced stuff.
Spec
A written description of what you’re building and what it should do. The source of truth for the project.
Diff
A side-by-side view of what changed in a file. Green lines are added, red lines are removed. Claude shows you one each time it edits.
Bash
The language used in Terminal. When Claude says it’s “running a Bash command”, it just means typing something in the terminal.
Bug
Something broken. The code doesn’t do what you expected. Fixing bugs is just detective work.
Superset
The Mac app that runs Claude Code in a calm, visual way. One window per project, files on the left, chat on the right. You can even run several agents at once.
Workspace
A single project inside Superset. Its own folder, its own isolated work area. Switching projects is just clicking a different tab.
Worktree
A sealed-off copy of your project on your Mac. Lets two things happen at once without collision. Superset uses them automatically.
Branch
A side-copy of your project you can experiment on. Like a tracing-paper study over a working drawing. Merge it back if it worked.
Push
Uploading your commits to GitHub. Makes them visible to others and safe from a lost laptop.
Pull Request (PR)
A formal proposal on GitHub to fold your branch into the main version. Reviewers comment, you refine, then everyone agrees before it merges.
Merge
The act of taking a branch and folding its changes into the main version. Usually happens at the end of a pull request.
LTS
Stands for “Long Term Support”. The steady, well-tested version of a tool. Always pick the LTS option when downloading.

End of Part 1.

You know more than enough to ship the first version. The rest you’ll pick up by doing.

Build the widget. Break it. Ask silly questions. Change your mind. That’s the whole job.

When you’re ready, open Part 2: the Ship It Graduate Deck.

Open Part 2  →

Made with for Janika.

Now go and ship something.