ShopMind — AI Shopping Agent

Kasparro Hackathon 2026 | Track 1

ShopMind replaces the browse-search-filter-compare loop with a single intelligent conversation. Tell it what you need. It asks smart questions. It recommends the right product with clear reasoning.

---

Demo Video

🎥 Watch Demo

---

Problem

Online shopping forces users through a painful loop: search → browse → filter → compare → still feel uncertain. Most users don't know exactly what to search for — they know what they need. ShopMind bridges that gap with conversation.

Solution

A conversational AI shopping agent that:

• Understands natural language intent ("I need a gift for my dad who exercises")
• Asks smart clarifying questions (budget, occasion, preferences)
• Fetches relevant products from Shopify
• Recommends 2-3 options with explicit tradeoffs
• Explains why it recommends what it recommends

---

Tech Stack

• Frontend: Next.js 16 + React + Tailwind CSS
• Backend: Next.js API Routes (server-side)
• AI: Google Gemini 2.5 Flash
• Products: Shopify Admin API (GraphQL)
• Language: TypeScript

---

Setup Instructions

Prerequisites

• Node.js v18+
• A Shopify Partner account with a development store
• A Google AI Studio account (free)

1. Clone the repo

git clone https://github.com/adyapathak22/shopmind-agent
cd shopmind-agent

2. Install dependencies

npm install

3. Set up environment variables

Create a .env.local file in the root directory:

GEMINI_API_KEY=your_gemini_api_key_from_aistudio.google.com
SHOPIFY_STORE_DOMAIN=your-store.myshopify.com
SHOPIFY_ADMIN_TOKEN=your_shopify_admin_api_token

Getting your Gemini API key:

1. Go to aistudio.google.com
2. Click "Get API Key" → "Create API key"
3. Copy the key

Getting your Shopify Admin token:

1. Go to your Shopify dev store admin
2. Settings → Apps and sales channels → Develop apps
3. Create an app → Configure Admin API scopes → enable read_products
4. Install app → API credentials → copy Admin API access token

4. Run the development server

npm run dev

5. Open in browser

Navigate to http://localhost:3000

---

How to Use

1. Type what you're looking for in plain language
2. Answer the agent's clarifying questions
3. Review the recommendations and tradeoffs
4. Ask follow-up questions if needed

Example queries:

• "I need shoes for a formal office environment"
• "Help me find a gift for someone who loves fitness, budget around $50"
• "I need something warm for winter, I'm going camping"
• "Show me something stylish but comfortable for everyday use"

---

Repository Structure

shopmind-agent/
├── app/
│   ├── api/
│   │   ├── chat/route.ts       # AI conversation endpoint (Gemini)
│   │   └── products/route.ts   # Shopify product fetcher
│   ├── globals.css
│   ├── layout.tsx
│   └── page.tsx                # Main chat UI
├── docs/
│   ├── product-doc.md          # Product thinking & decisions
│   ├── technical-doc.md        # Architecture & implementation
│   └── decision-log.md         # Running log of key decisions
├── .gitignore
├── next.config.ts
├── package.json
└── README.md

---

Screenshots

---

Contribution Note

Team: Adya Pathak & Rupali Pasa | KIIT University, CSE Branch

This was a collaborative build where both team members contributed equally across product thinking and engineering.

Adya Pathak:

• Frontend development (chat UI, message rendering, UX design)
• AI layer integration (Gemini API, conversation prompt engineering)
• Product document and Technical document
• Demo video narration

Rupali Pasa:

• Backend development (Next.js API routes)
• Shopify Admin API integration (GraphQL queries, product fetching)
• Architecture design and user journey design
• Decision log and scope decisions

Joint contributions:

• Product concept and problem framing
• Conversation flow design
• Testing and debugging
• README and submission preparation

Both team members spent approximately equal time on product thinking and engineering work throughout the hackathon.

---

Documents

Product Document Technical Document Decision Log