Skip to content

README.md

Latest commit

 

History

History
128 lines (99 loc) · 4.22 KB

README.md

File metadata and controls

128 lines (99 loc) · 4.22 KB
Raw

SkinLab

AI-powered skin analysis and skincare recommendation iOS app.

Overview

SkinLab helps users understand their skin health through AI-powered analysis and track skincare effectiveness over a 28-day verification cycle. The app focuses on data-driven insights that users want to share.

Core Philosophy: Make skincare effects visible ("让护肤效果看得见")

Target Users: 18-35 year olds focused on skincare

Differentiation:

  • Effect verification engine with 28-day tracking cycles
  • Skin twin matching for personalized recommendations
  • Anti-ad commitment - evidence-based suggestions only

Features (Completed)

fn-2: Engagement (Daily Streaks & Achievement Badges)

  • Daily streak tracking for check-ins with longest streak display
  • Achievement badges and dashboard for milestones
  • Milestone celebration animations (respects reduced motion)
  • Streak freeze mechanic (1 per 30 days)
  • Local notifications for streak reminders and at-risk warnings

fn-3: Photo Standardization & Lifestyle Correlation

  • Day 0 baseline creation to start tracking sessions from analysis
  • Standardized photo capture guidance with real-time feedback
  • Photo quality and reliability scoring at capture time
  • Lifestyle inputs are optional and only saved when explicitly set
  • Lifestyle correlation insights based on real score deltas

Tech Stack

Component Technology
Platform iOS 17+
Language Swift 5.9+
UI Framework SwiftUI
Architecture MVVM + Clean Architecture
Storage SwiftData
AI Gemini 3.0 Flash Vision API
Image Processing Vision Framework

Getting Started

Requirements

  • Xcode 15.0+
  • iOS 17.0+ deployment target
  • Swift 5.9+
  • Gemini API key (see Secrets.xcconfig.template)

Setup

  1. Clone and open the project:

    git clone <repository-url>
cd SkinLab
open SkinLab.xcodeproj

  2. Configure secrets:

    • Copy Secrets.xcconfig.template to Secrets.xcconfig
    • Set your API key in the new file (see template for key name)
    • Do NOT commit Secrets.xcconfig to version control

  3. Build and run on simulator or device:

    xcodebuild -scheme SkinLab -destination 'platform=iOS Simulator,name=iPhone 15'

Running Tests

xcodebuild test -scheme SkinLab -destination 'platform=iOS Simulator,name=iPhone 15'

Project Structure

SkinLab/                    # App source code
├── App/                    # App entry point and configuration
├── Core/
│   ├── Network/            # API services (GeminiService)
│   └── Utils/              # Extensions and utilities
├── Features/
│   ├── Analysis/           # AI skin analysis feature
│   ├── Tracking/           # 28-day effect tracking
│   ├── Engagement/         # Streaks and achievements
│   ├── Community/          # Skin twin and sharing
│   ├── Products/           # Product recommendations
│   ├── Profile/            # User profile management
│   ├── Celebration/        # Achievement celebrations
│   └── Sharing/            # Social sharing services
├── Services/               # Shared business logic services
├── UI/
│   ├── Components/         # Reusable UI components
│   └── Theme/              # Design system and styling
└── Resources/              # Assets and data files

SkinLabTests/               # XCTest target for unit tests
.flow/                      # Flow-Next specs and tasks

Planning Artifacts

The .flow/ directory contains project planning and task tracking:

  • Epic specifications in .flow/specs/
  • Task breakdowns in .flow/tasks/
  • See .flow/usage.md for workflow details

Key planning documents:

  • fn-1: Competitor analysis and improvement roadmap
  • fn-2: Engagement features (streaks, badges)
  • fn-3: Photo standardization and lifestyle correlation
  • fn-4: Code quality and test coverage

Contributing

See CLAUDE.md for development guidelines and AI agent instructions.

Key conventions:

  • Use Flow-Next (.flow/bin/flowctl) for task tracking
  • Follow MVVM + Clean Architecture patterns
  • Write unit tests for core services
  • Keep views under 600 lines