PriceHunt MCP
MCP-powered AI tool that finds the cheapest 4+ star rated products across Daraz, Telemart, and iShopping. Uses Gemini + LangChain for smart product matching with a Streamlit chat interface.
Ask AI about PriceHunt MCP
Powered by Claude ยท Grounded in docs
I know everything about PriceHunt MCP. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
๐๏ธ PriceHunt
A Model Context Protocol (MCP) implementation that finds the lowest-priced products with good ratings (4+ stars) across major Pakistani e-commerce platforms including Daraz, Telemart, and iShopping.
MCP
Model Context Protocol (MCP) is a standardized protocol that enables AI applications to securely connect to external data sources and tools. It acts as a bridge between AI models (like Gemini) and various services, databases, APIs, and applications.
MCP Architecture Components:
- MCP Servers - Provide specific tools, resources, or data to clients
- MCP Clients - AI applications that want to access external resources
- Transport Layer - Communication mechanism between clients and servers
๐ฏ Project Overview
This project demonstrates MCP implementation by creating:
- MCP Server: Provides three tools for scraping Pakistani e-commerce sites
- MCP Client: Uses LangChain + Google Gemini to orchestrate tool calls
- Streamlit Frontend: User-friendly web interface for product searches
Note: In this project both server and client run on the same host for learning purposes.
โจ Features
- ๐ Multi-Platform Search: Scrapes Daraz, Telemart, and iShopping simultaneously
- โญ Quality Filtering: Prioritizes products with 4+ star ratings
- ๐ฐ Price Search: Finds the lowest-priced genuine products
- ๐ค AI-Powered: Uses Google Gemini for intelligent product matching
- ๐ฌ Chat Interface: Conversational UI with memory
- ๐ Async Processing: Non-blocking operations for better performance
๐๏ธ MCP Architecture
This Project's MCP Implementation:
โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ
โ Streamlit โ โ MCP Client โ โ MCP Server โ
โ Frontend โโโโโบโ (LangChain + โโโโโบโ (FastMCP) โ
โ (app.py) โ โ Gemini) โ โ โ
โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโ
โ
โผ
โโโโโโโโโโโโโโโโโโโ
โ E-commerce โ
โ Websites โ
โ โข Daraz.pk โ
โ โข Telemart.pk โ
โ โข iShopping.pk โ
โโโโโโโโโโโโโโโโโโโ
MCP Tools Defined:
get_daraz_products(query)- Scrapes Daraz with 4+ rating filterget_telemart_products(query)- Scrapes Telemart search resultsget_ishopping_products(query)- Scrapes iShopping catalog
๐ Project Structure
PriceHunt-MCP/
โโโ project/ # Client-side code
โ โโโ app.py # Streamlit web interface
| โโโ mcp_client.py # MCP Client with LangChain integration
| โโโ mcp_server.py # MCP Server with 3 e-commerce tools
โโโ python-version # Python version specification
โโโ pyproject.toml # Python project configuration
โโโ README.md # This file
โโโ uv.lock # UV dependency lock file
๐ Installation & Setup
1. Clone the Repository
git clone https://github.com/FassihShah/PriceHunt-MCP.git
cd PriceHunt-MCP
2. Create Virtual Environment
# Create virtual environment
python -m venv venv
# Activate virtual environment
venv\Scripts\activate
3. Install Dependencies
Since we're using uv, install dependencies with:
# If using uv (recommended)
uv install
# Or using pip with requirements.txt
pip install -r requirements.txt
If you don't have uv installed:
# Install uv first
pip install uv
# Then install dependencies
uv install
4. Set Up Environment Variables
Create a .env file in the project root:
GOOGLE_API_KEY=your_google_gemini_api_key_here
๐ฅ๏ธ Using with Claude Desktop
This MCP server can also be integrated directly with Claude Desktop application, allowing to use the e-commerce tools directly in your conversations with Claude!
Setup for Claude Desktop:
1. Install Claude Desktop:
- Download from Claude Desktop
- Make sure you have the latest version
2. Configure Claude Desktop: Open the Claude Desktop configuration file:
Windows:
code %APPDATA%\Claude\claude_desktop_config.json
3. Add Your MCP Server:
Create or update the claude_desktop_config.json file:
{
"mcpServers": {
"ecommerce-scraper": {
"command": "python",
"args": ["/path/to/your/project/mcp_server.py"],
"env": {
"PYTHONPATH": "/path/to/your/project"
}
}
}
}
Once configured, you can directly ask Claude things like:
- "Find me the cheapest Ronin Earbuds"
Claude will automatically use these MCP tools to scrape the websites and provide results!
๐ฎ Usage
Method 1: Claude Desktop Integration
After setting up Claude Desktop configuration (see section above)
Method 2: Streamlit Web Interface
streamlit run app.py
Method 3: MCP Inspector (Development & Testing)
Use the official MCP Inspector to test and debug your server:
uv run mcp dev mcp_server.py
This will:
- Launch a web interface
- Test all your tools interactively
- View tool schemas and parameters
๐ Learning Outcomes
This project demonstrates:
- MCP Protocol: Understanding of server/client architecture
- AI Integration: LangChain + LLM tool orchestration
- Async Programming: Non-blocking operations
