wardrobe
Your clothes, extracted and organized with gpt-image.
What is it?
What it is
Wardrobe is an open‑source application that uses OpenAI’s GPT‑image and Responses APIs to detect garments in photos, extract clean cutouts, generate modeled previews, and organize a personal clothing library locally.
Why it exists
To give users a simple, private way to import their own clothing photos, create modeled images of each item, and generate outfit ideas without relying on external services, keeping all data on their machine.
Who should use it
Developers, fashion enthusiasts, and users seeking AI-powered wardrobe organization. Agents or services managing user clothing data.
Who should avoid it
Non-technical users, those without an OpenAI API key, or individuals unwilling to provide a model-reference image.
How it works
A quick walkthrough in plain English
How wardrobe works
Step 1 of 3
You interact with it
Open wardrobe, send a request, or connect it to your stack.
Features
Advantages
- MIT license permits free use and modification
- Node.js/JavaScript stack is widely accessible
- Local data storage ensures privacy and offline access
- Simple quick-start with npm and dev server
- Integration with OpenAI's latest vision and image models
- Agent-friendly via Codex skills for automation
- Extensible configuration via environment variables
Disadvantages
- Requires a valid OpenAI API key (incurs usage costs)
- Needs a model-reference PNG for personalized results
- Dependent on external OpenAI APIs; service changes or outages affect functionality
- Limited to environments with Node 22+ and npm
- No built-in support for other AI providers or offline models
- Documentation assumes familiarity with OpenAI API and Codex
Installation
native
git clone https://github.com/tandpfun/wardrobe.git cd wardrobe npm install cp .env.example .env npm run dev
FAQ
How do I get started with Wardrobe?
Clone the repository, run `npm install`, copy `.env.example` to `.env`, and run `npm run dev`. You can then access the application at localhost:5173.
Why is the importer disabled?
The importer remains disabled until you provide an `OPENAI_API_KEY` in your `.env` file and place a PNG reference photo of yourself at `data/model-reference.png`.
What are the Codex skills available in this repository?
Wardrobe includes two Codex skills: one for importing clothes and generating modeled item photos, and another for styling complete outfits and generating a modeled lookbook.
Where is my data stored?
All original images, processing jobs, generated images, and the JSON database are stored locally within the `data/` directory.
How does the image extraction process work?
The application uses the OpenAI Responses API to detect garments in a photo and the OpenAI Images API to extract clean product cutouts.
Can I customize the AI models used?
Yes, you can configure the `OPENAI_VISION_MODEL`, `OPENAI_IMAGE_MODEL`, and `OPENAI_IMAGE_QUALITY` via environment variables.
Featured in Videos
YouTube tutorials and walkthroughs for wardrobe
Alternatives
Similar projects ranked by category, topics, and text overlap.
