A Pikmin Bloom companion app for my fellow collectors
// Build Note
PikGarden is a fan-made tracker for Pikmin Bloom’s decor Pikmin collection. It helps players track what they’ve collected and find nearby locations for what they’re still missing.
Timeline
One week to launch. Ongoing iteration.
Stack
Vanilla JS · Firebase Auth · Firestore · OpenStreetMap · Vercel · Claude Code
Status
Live · 1,500+ players · Updated monthly
PikGarden is not affiliated with Niantic or Nintendo. Visuals shown are original or recreated work.
Where it started
I started playing Pikmin Bloom as a gentle reason to go outside more. It woke up the completionist in me. With so many decor types and stages, tracking progress got hard fast. Some players were building their own spreadsheets. I wanted something that worked across devices and helped me find what I was missing, not just record what I already had.
I sketched the core concept, then opened Claude and started building.
↪ My original message to Claude, the starting point for everything that followed.
What it became
Think of a teacher’s roll book, but for your little ones.
PikGarden has two sides. The Garden Tracker is the attendance sheet: a card for every decor type, with a progress bar showing how far along you are. Mark Pikmin individually by type and color, or check off a whole category at once. A stats view brings your collection together.
The Decor Finder is for the ones who haven’t shown up. Confirm your location, choose a decor type, and browse nearby places in a list. Switch to the map when you want one, then open Google Maps for directions to your next stop.
Find a seedling, grow the new guy, and mark them present on the attendance sheet.
PikGarden runs in the browser, and signing in keeps your collection in sync across devices. Since launching in April, more than 1,500 players have used it, with an average session of over 4 minutes.
I update it monthly as new decor and events arrive, and keep adding languages on my own time.
The judgment calls
Building with Claude took repeated steering and iteration. I shaped the product logic, defined the interactions, and designed the visual details, while Claude implemented the code.
Here are a few decisions that shaped the result.
#1
Designing for how a collector actually thinks
Claude’s first pass put “Spot finder” before “My collection.” The features were there, but the priority was backwards for how I wanted the app to work. You need to know what you have before you can meaningfully search for what you don’t.
I walked Claude through how a collector would open the app and decide what to do next, then had it change the tab priority so collection came first.
#2
Recreating what players already know
Claude chose emojis for the category icons because they were the fast, easy choice. They were functional, but generic and unfamiliar. I wanted the icons to match the visual language players already knew from the game to lower recognition friction.
I mapped out which decor types could stay as emojis and which needed custom artwork. Then I redrew the icons and medals in Figma. Claude handled integrating them into the interface.
#3
Adding friction on purpose
I didn’t want the Decor Finder to open by loading everything onto a map. That felt like the wrong default, both technically and behaviorally. Before showing places, the app needed to understand what the collector was looking for.
I defined a deliberate sequence: confirm location, choose a decor type, see nearby places in a list, then switch to the map if needed. Those extra steps gave the search a focus rather than presenting everything at once.
I broke the flow into smaller tasks and worked through them with Claude step by step, rather than trying to specify the whole experience in one prompt.
For directions, I chose to hand off to Google Maps. PikGarden’s job was to help players find their next stop, then let them continue in a familiar navigation tool.
#4
Small moments, made on purpose
I put care into details people may or may not notice.
Pikmin-shaped checkboxes that I drew instead of default checkmarks. Monthly events pinned at the top so players could find them at a glance. A small celebration when someone finishes a collection. A stats view for fellow collection nerds.
I designed each of these, then worked with Claude to build them. Some are practical. Others are there to make the app feel a little more cared for.
None was strictly required. I added them because I wanted to.
Process
First time building this way
PikGarden was my first project built with Claude. The workflow looked like this:
Think through the idea with Claude → Build a working version → Refine visual details in Figma → Back into code → Repeat
The working version gave me something concrete to react to. I could see where the flow felt wrong, which details needed my own design work, and what to tackle next. It wasn’t a single prompt followed by a finished app, but a loop of giving context, reviewing the result, and refining it.
I work differently now, and each project since has pulled me toward a different workflow.
