Can I vibecode VistaCreate?

Build a personal replacement for VistaCreate in an empty repository.
Use Next.js 15 (App Router), TypeScript, React, Tailwind CSS, fabric.js (for canvas manipulation), and IndexedDB (via localforage); do not offer alternative stacks.

DATA MODEL:
- `Project`: id, title, width, height, canvas_json, thumbnail_data_url, created_at, updated_at
- `BrandKit`: id, name, colors (json array), font_urls (json array), logo_urls (json array)
- `Asset`: id, type (image, vector), url, category, source

USER JOURNEYS & SCREENS:
1. **Home/Template Selector (`/`)**: A grid displaying predefined template sizes (e.g., Instagram Post 1080x1080, Facebook Cover). Clicking one creates a new `Project` and navigates to the Editor.
2. **Design Editor (`/editor/{id}`)**: A complex UI with a central Fabric.js canvas. Left sidebar has tabs: 'Templates', 'Assets', 'Text', 'Brand Kit', 'Uploads'. Top toolbar changes contextually based on the selected object (e.g., color picker, font dropdown, opacity slider).
3. **Brand Kit Manager (`/brand`)**: A simple form interface to upload logos, define primary/secondary hex colors, and upload custom font files (TTF/WOFF) saving to `BrandKit`.
4. **Export Modal**: Triggered from the Editor, allows downloading the canvas as a PNG, JPG, or PDF. 

TECHNICAL SPECIFICATIONS:
- Canvas Engine: Use `fabric.js` to handle complex rendering, grouping, z-index, SVG importing, and text manipulation. The canvas state should be serialized to JSON (`canvas_json`) and saved to IndexedDB on every modification.
- Image Processing: Implement a basic "Remove Background" tool by integrating an external API (like remove.bg or an open-source model via Hugging Face Inference API) that returns a transparent PNG.
- Font Loading: Use FontFace API to dynamically load custom fonts from the `BrandKit` into the document before rendering them on the canvas.
- Asset Library: Seed a local SQLite database (for the desktop/server environment) with a curated list of royalty-free Unsplash images and SVG icons.

Make the first run work locally with one documented command.
Store all user data locally.
Put API keys in .env, ship .env.example.
Implement undo, redo, and autosave.
Include clear empty, loading, success, and recoverable error states.
Write focused tests for the canvas serialization and export functions.
Create a README with setup, architecture, and backup steps.
Do not add commercial stock libraries or enterprise approval workflows.