ADHD-Closet

Frequently Asked Questions (FAQ)

Common questions about Wardrobe AI Closet.

General Questions

What is Wardrobe AI Closet?

A self-hosted, single-user wardrobe organizer powered by AI. It helps you catalog your clothes, generate outfit suggestions, and reduce decision paralysis. Built with ADHD users in mind.

Who is this for?

Anyone who:

Why self-hosted?

Is this free?

Feature Questions

What can the AI do?

  1. Generate catalog images: Clean, professional-looking photos from your casual pics
  2. Infer metadata: Automatically detect category, colors, and attributes
  3. Extract labels: Read brand, size, and material info from label photos
  4. Suggest outfits: Generate 1-5 outfit combinations based on weather, vibe, occasion

Can I use it without AI?

Yes! AI features are optional. You can:

You’ll miss out on time-saving automation, but the app is fully functional without AI.

Does it track what I wear?

Not automatically in v1. You can manually rate outfits (👍/👎/😐) after wearing them, which influences future suggestions.

Can I add items without photos?

No, photos are required. The app is built around visual organization since that’s more effective for ADHD users.

How many items can I store?

Tested with 10,000+ items. Limits depend on:

Can I organize items into collections?

Not custom collections in v1. You can:

Does it work offline?

The app works offline for browsing and manual edits. AI features require internet (calls OpenRouter API).

Can multiple people use one installation?

It’s designed as single-user. For multiple users:

Technical Questions

What platforms does it run on?

What are the system requirements?

Minimum:

Recommended:

Do I need to know how to code?

Not for basic usage. You’ll need command-line skills for:

For customization: Basic TypeScript/React knowledge helpful.

Can I run this on a Raspberry Pi?

Yes, but:

Does it work on mobile?

Yes! The web interface is fully responsive:

There’s no native iOS/Android app, but you can “Add to Home Screen” for app-like experience.

Deployment Questions

How do I install it?

See DEPLOYMENT.md for full guides:

Can I access it from anywhere?

Yes, if you:

  1. Deploy to a VPS with public IP
  2. Setup domain name (optional but recommended)
  3. Enable HTTPS (certbot, Let’s Encrypt, or Cloudflare)
  4. Secure it: Add authentication (not built-in!)

Warning: Default installation has NO authentication. Don’t expose to internet without protection.

How do I update to the latest version?

# Docker
git pull
docker-compose up -d --build

# VPS
git pull
cd app
bun install
bun run build
sudo systemctl restart wardrobe

How do I backup my data?

Three methods:

  1. App export: Settings → Export Wardrobe (downloads ZIP)
  2. Database backup: pg_dump your PostgreSQL database
  3. File backup: Copy DATA_DIR/images directory

Recommended: Use all three, scheduled weekly.

What if I lose my data?

Restore from backup:

  1. Settings → Import Wardrobe → Upload ZIP
  2. Choose “Replace” mode to restore fully
  3. Or “Merge” mode to add to existing data

Cost Questions

How much does AI cost?

Depends on usage:

Typical costs per item:

Example monthly costs:

Cost control:

Which AI models should I use?

Recommended (balance of quality and cost):

Budget (cheapest):

Best Quality (expensive):

Configure in .env (see API_DOCUMENTATION.md).

Can I use local AI instead of OpenRouter?

Not in v1. You’d need to:

  1. Run local models (Ollama, LocalAI, etc.)
  2. Modify app/lib/ai/openrouter.ts to use local endpoint
  3. Test thoroughly (local models vary in quality)

Contributions welcome!

Design & UX Questions

Why is it designed this way?

Every design decision prioritizes ADHD users:

See SPEC.md section 7.1 for detailed UX rules.

Can I change the colors/theme?

Yes! Modify app/app/globals.css color tokens:

:root {
  --md-sys-color-primary: #6750A4;  /* Change to your color */
  /* ... */
}

Dark mode is built-in (Settings → Dark Mode).

Can I add more categories?

Yes, but requires code changes:

  1. Edit app/prisma/schema.prisma (add to Category enum)
  2. Run bunx prisma migrate dev
  3. Update CATEGORIES array in components

Why no drag-and-drop outfit builder?

It’s on the roadmap! Desktop power tools (≥1024px) will include:

Can I hide certain features?

Not yet. Future: Settings to disable features you don’t use.

Troubleshooting Questions

The AI isn’t working. What do I check?

  1. OpenRouter API key valid? (https://openrouter.ai/keys)
  2. Sufficient credits? (https://openrouter.ai/account)
  3. Redis running? (redis-cli ping)
  4. Check AI job status in item detail page
  5. See TROUBLESHOOTING.md

Images aren’t loading. Why?

Common causes:

  1. File permissions (app can’t read image directory)
  2. Wrong DATA_DIR in .env
  3. Disk full
  4. Corrupt image file

See TROUBLESHOOTING.md → “Image Upload Failures”

The app is slow. How do I fix it?

  1. Check available RAM: free -h
  2. Check disk I/O: iostat
  3. Optimize images (see TROUBLESHOOTING.md)
  4. Increase PostgreSQL shared_buffers
  5. Use SSD instead of HDD

How do I reset everything?

CAUTION: Deletes all data!

# Stop app
sudo systemctl stop wardrobe

# Drop and recreate database
sudo -u postgres psql
DROP DATABASE wardrobe;
CREATE DATABASE wardrobe OWNER wardrobe;
\q

# Clear Redis
redis-cli FLUSHALL

# Delete images
rm -rf /var/www/wardrobe/data/images/*

# Reinitialize
cd /var/www/wardrobe/app
bunx prisma migrate deploy
sudo systemctl start wardrobe

Privacy & Security Questions

Is my data private?

Yes! Everything stays on your server:

What data does OpenRouter see?

Only when you use AI features:

OpenRouter privacy: https://openrouter.ai/privacy

Should I password-protect it?

YES if accessible from internet. Options:

  1. Nginx basic auth (simplest)
  2. Tailscale/ZeroTier (VPN, most secure)
  3. Cloudflare Access (SSO)
  4. Custom auth (add to app, see CONTRIBUTING.md)

Can I use this for commercial purposes?

The app is open source (MIT license). You can:

But you must:

Contributing Questions

How can I contribute?

See CONTRIBUTING.md for:

I found a bug. What do I do?

  1. Check TROUBLESHOOTING.md
  2. Search existing GitHub issues
  3. Open a new issue with:
    • Steps to reproduce
    • Expected vs actual behavior
    • Screenshots (if UI bug)
    • Environment (OS, Node version, etc.)

I have a feature idea!

Open a GitHub Discussion or Issue:

Can I translate the app?

Not yet internationalized. If you want to add i18n:

  1. Open a discussion to coordinate
  2. Use next-intl or react-i18next
  3. Follow Material Design 3 localization guidelines

Roadmap Questions

What features are planned?

High Priority:

Medium Priority:

Low Priority (or community-driven):

When is v2.0 coming?

No timeline. This is a passion project. Contributions welcome to accelerate development!

Will there be a mobile app?

Not planned. The web interface is fully responsive and works great as a PWA (Add to Home Screen).

Native apps require:

Still Have Questions?

Happy organizing! 👗👔🧥