Full Stack MCP Stock Analyser

Stock Analyzer using MCP Server + AI

AI-powered stock analysis platform that connects financial market data with intelligent natural language insights using OpenAI GPT. Built with a FastAPI MCP server backend and a Next.js frontend, this system analyzes stock trends, displays financial metrics, and generates downloadable PDF stock reports with AI summaries.

⚡ Quick Start Guide

✅ Start Backend (FastAPI MCP Server)

cd stock_analyzer
python3 -m venv venv
source venv/bin/activate      # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn mcp_server.server:app --reload --port 8000

✅ Start Frontend (Next.js UI)

cd frontend
npm install
npm run dev

Now open your browser ➝ http://localhost:3000

✅ Features

🏗️ Architecture Overview

Frontend (Next.js UI)
        │
        ▼
Backend API (FastAPI MCP Server)
  ├─ /analyze_stock
  ├─ /download_report
        │
 ┌──────┴─────────┐
 │                │
Yahoo Finance   OpenAI GPT

📦 Project Directory Structure

stock_analyzer/
├── analyzer.py                # Stock analysis logic
├── llm_reporter.py            # AI summary using GPT
├── report_pdf.py              # PDF report generator
│
├── mcp_server/
│   ├── server.py              # FastAPI backend
│   ├── templates.py           # Prompt templates
│   ├── traders.py             # Buy/Hold/Sell logic
│   ├── mcp_params.py          # Config
│
├── frontend/                  # Next.js frontend
│   ├── app/page.tsx
│   ├── components/AnalyzerForm.tsx
│   ├── package.json
│
├── reports/                   # Generated PDFs + charts
├── .env                       # Environment file

stock_analyzer/
│
├── analyzer.py
├── llm_reporter.py
├── main.py
│
├── mcp_server/
│   ├── __init__.py
│   ├── server.py               # UPDATED (adds CORS + /download_report)
│   ├── mcp_params.py
│   ├── templates.py
│   └── traders.py
│
├── report_pdf.py               # NEW (PDF generation helper)
├── reports/                    # charts + cached PDFs
│
├── requirements.txt            # ensure: fastapi, uvicorn, reportlab
├── .env
│
└── frontend/                   # NEW Next.js 14 app
    ├── package.json
    ├── next.config.js
    ├── postcss.config.js
    ├── tailwind.config.ts
    ├── tsconfig.json
    ├── app/
    │   ├── globals.css
    │   └── page.tsx
    └── components/
        └── AnalyzerForm.tsx
└── README.md

🔧 Environment Setup

Create .env file inside project root:

OPENAI_API_KEY=your_openai_api_key
DEFAULT_STOCK=RELIANCE.NS
DEFAULT_PERIOD=3mo

🔌 API Endpoints (Backend)

Example request:

http://localhost:8000/analyze_stock?symbol=TCS.NS&period=6mo

📄 Sample PDF Output

✅ AI Summary ✅ Stock Chart ✅ Metrics Table ✅ Time Period ✅ Downloadable & Shareable Report

📎 *Reports are saved inside /reports*

🚀 Deployment

Deploy Backend

You can deploy the FastAPI server to:

• Render.com
• Railway.app
• Azure App Service
• AWS EC2 or Lightsail

Deploy Frontend (Recommended)

npm install -g vercel
vercel deploy

Update API URL:

NEXT_PUBLIC_API_BASE=https://your-api-url.com

🛠️ Troubleshooting

✅ Roadmap

• [ ] Deploy on Vercel + Render
• [ ] Add multi-stock comparison
• [ ] Add Bollinger Bands & MACD
• [ ] Add database for user history
• [ ] Add authentication
• [ ] Host as AI microservice

👨‍💻 Author

Built by Anish Roychowdhury For AI engineering, education & financial data insights.

📜 License

MIT License – Free to use, learn & modify.

Full-Stack-MCP-Stock-Analyser