.cursorrule.md
Purpose
This file provides Cursor AI with guidelines and context for generating code and configuring the Voice-Driven Photo & Document Manager. It outlines the project scope, folder structure, dependencies, and specific prompts to ensure Cursor’s AI functions optimally for this project.
Project Overview
- Project Name: Voice-Driven Photo & Document Manager
- Goal:
- Create an iOS app (Swift/SwiftUI) for voice-powered photo capture and metadata tagging.
- Develop a React-based web dashboard to view, search, and manage uploaded photos and documents.
- Use Firebase for data storage, authentication (optional), and file storage.
- Primary Features:
- Voice Command Parsing (iOS):
- “Take photo of Floor 2, Unit 205” → tags photo with
floor=2, unit=205.
- “Show me photos of Floor 1” → retrieves all photos with
floor=1.
- Photo Storage:
- Upload photos to Firebase Storage, store metadata in Firestore.
- Web Dashboard:
- Display a list or gallery of tagged photos.
- Basic search functionality (by floor, unit, or custom tags).
- Optional Document Linking:
- Attach PDFs or drawings to specific floors/units.
Guidelines for Cursor AI
-
Language & Framework Prompts
- Swift/SwiftUI for iOS:
- Use Apple’s Speech Framework for voice recognition.
- Integrate Firebase SDK (Core, Firestore, Storage).
- React for Web:
- Use the
create-react-app or Vite template.
- Integrate Firebase (Firestore & Storage) using
firebase npm package.
-
Code Generation Approach
- Provide concise and commented code snippets.
- Whenever user prompts for code, generate self-contained functions or components that are easy to copy into the project.
- Avoid extraneous logic or advanced patterns beyond the user’s scope (the user is a beginner).
-
Directory Structure
- iOS (Swift) Project:
VoicePhotoManager-iOS/
├─ VoicePhotoManager.xcodeproj
├─ VoicePhotoManager/
│ ├─ App.swift
│ ├─ ContentView.swift
│ └─ FirebaseConfig.swift (for Firebase init)
└─ Podfile
- Web (React) Project:
VoicePhotoManager-Web/
├─ src/
│ ├─ App.js
│ ├─ firebase.js (Firebase init)
│ ├─ components/
│ └─ pages/
├─ package.json
└─ .gitignore
-
Required Dependencies
- iOS (in
Podfile):
pod 'Firebase/Core'
pod 'Firebase/Firestore'
pod 'Firebase/Storage'
pod 'Firebase/Auth' # if authentication is needed
- Web (
package.json):
{
"dependencies": {
"firebase": "^<latest_version>",
"react": "^<latest_version>",
"react-dom": "^<latest_version>"
}
}
-
Sample Prompts for Cursor
- Voice Recognition Setup (Swift)
// Prompt: "Generate Swift code using Apple's Speech framework
// to transcribe short voice commands. Return the recognized text."
- Firebase Upload Function (Swift)
// Prompt: "Generate Swift code to upload an image to Firebase Storage
// and save its metadata (floor, unit) in Firestore."
- React Photo Gallery
// Prompt: "Generate a React component named 'PhotoGallery' that fetches
// photo metadata from Firestore, displays images, and shows floor/unit tags."
- Voice Search (React)
// Prompt: "Generate a React hook that uses the Web Speech API
// to capture voice input, convert to text, and filter a photo list accordingly."
-
Implementation Details
- Transcription Parsing:
- Simple approach: parse recognized text for keywords like “Floor” and “Unit.”
- Example: "Floor 2, Unit 205" →
floor=2, unit=205.
- CRUD Operations:
- iOS and React should be able to:
- Create (upload photos with metadata).
- Read (display items).
- Update (change tags if needed).
- Delete (remove photos or documents).
- Testing & Debugging:
- Use the iOS simulator for basic tests (voice recognition might require a physical device).
- For React, run locally with
npm start or yarn start.
-
Styling & UI
- Keep the UI simple and intuitive (the user is a beginner).
- SwiftUI for iOS handles layout nicely; for React, basic CSS or a library like Tailwind can be used if desired.
-
Deployment
- iOS: Deploy using TestFlight or direct device debugging.
- Web: Deploy on Firebase Hosting or any static hosting provider.
Additional Notes
- Keep Code Simple: The user is new to programming, so avoid overly abstract or advanced patterns.
- Always Provide Comments: Each generated snippet should include short comments explaining key lines of code.
- Focus on Core Requirements: Priority is voice capture → tagging → uploading → searching. Anything else can be added later.
When receiving user prompts in Cursor, ensure to:
Output minimal, well-organized code.
Include inline comments for clarity.
Provide short usage instructions if needed.