Skip to content

Akkipr/acari

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Acari 🚗

An AI-powered car modification companion for SwiftUI/iOS. Acari helps car owners plan, budget, and visualize mods for their specific vehicle — and find real-world help (mechanics, detailers, tow trucks) when they need it.

Built for a Hack the Valley - Hackdays hackathon; powered by Firebase Auth, Google's Gemini API (text, image, and Veo video), and the Google Places API.


Features

Account & Profile

  • Onboarding flow that captures the user's car (make, model, year, colors, mod budget) and generates a custom AI sticker/decal of their exact vehicle for the dashboard

Dashboard

  • At-a-glance budget summary card (spent / planned / remaining)
  • Quick access to every feature below, tailored to the user's car make

Mod Visualizer

  • Describe a mod — pick a preset (widebody kit, lowered + black rims, matte wrap, racing livery) or type a custom prompt. You will get image-edits of your own car photo
  • Upload a part — upload a photo of a real part. Composites it realistically onto the user's car instead of guessing a generic version

Cinematic Reveal Video

  • Upload 2–3 photos of the car and pick a backdrop theme (sunset highway, Tokyo night, track day, etc.)
  • Veo 3.1 generates a short cinematic clip: two slow pans across the parked car, then a hard cut to it driving through the chosen scene
  • Save directly to Photos

Garage Assistant

  • Gemini-powered chatbot scoped to the user's specific make/model
  • Gives step-by-step repair instructions for DIY-friendly jobs, and flags when a job needs a professional

Find Services

A single hub covering four categories:

Category What it does
Mechanical Work Pick a service (oil change, brakes, general service, etc.) → Get an estimated cost & time → Google Places surfaces top-rated shops nearby
Detailing Same flow for exterior wash, interior, full detail, ceramic coating
Emergency Services Towing, jump starts, lockouts, flat tires — wider search radius, faster path to help
Maintenance Log Fully local tracker (no AI) — log tire/brake/rotor/fluid changes with date, odometer reading, cost, and notes; see "last done" at a glance

💰 Budget Tracker

  • Manually add parts you've already found (with photo + cost)
  • Ask Gemini where to buy a specific part, scoped to your car — one tap adds it to your list
  • Track spent vs. planned vs. budget with progress visualization

Tech Stack

  • UI: SwiftUI (iOS)
  • Auth: Firebase Authentication (Email/Password)
  • AI: Google Gemini API
    • gemini-3.6-flash — chat, cost/time estimates, part suggestions (structured JSON)
    • gemini-3.1-flash-image-preview — sticker generation, mod visualization, part compositing
    • veo-3.1-generate-preview — cinematic reveal video generation
  • Location/Places: Google Places API (New) — Text Search, biased by geocoded location

Getting Started

Prerequisites

  • Xcode 16+
  • An iOS 26.2+ deployment target (adjust as needed)
  • A Google AI Studio API key (Gemini + Veo)
  • A Google Cloud project with Places API (New) enabled
  • A Firebase project with Email/Password auth enabled

1. Clone & open

git clone https://github.com/Akkipr/acari.git
open HackDays.xcodeproj

2. Add Firebase

  1. In Firebase console: Authentication → Sign-in method → enable Email/Password.
  2. Add an iOS app in the Firebase console, download GoogleService-Info.plist, and drag it into the Xcode project (checked for your app target).
  3. In Xcode: File → Add Package Dependencieshttps://github.com/firebase/firebase-ios-sdk → add the FirebaseAuth product.

3. Add API keys

Open Secrets.swift and set:

enum Secrets {
    static let geminiAPIKey = "your-key"
    static let googlePlacesAPIKey = "your-key"
}

⚠️ Do not ship these in a production client binary.

4. Build & run

Select a simulator or device and hit Run. On first launch you'll be prompted to sign in — create an account (or use a demo user you've set up in the Firebase console) and build your garage profile.

Notes & Limitations

  • Maintenance log distances/mileage are tracked in kilometers — swap to miles in MaintenanceRecord/MaintenanceTrackerView if needed.
  • Veo video generation typically takes 1–3 minutes and is billed per output second — be mindful of costs during testing.

License

Read full description here

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages