MBIQ — Portfolio Intelligence Platform

A comprehensive stock portfolio tracker and earnings call analysis tool built with React, Supabase, and AI-powered transcript analysis.

Tech Stack

• Frontend: React + Vite + TypeScript + Tailwind CSS + shadcn/ui
• Backend: Supabase (Postgres, Edge Functions, RLS)
• Charts: Recharts
• State: TanStack React Query

---

Local Development Setup

Prerequisites

• Node.js (v18+)
• Supabase CLI (npm install -g supabase)
• Docker (required for local Supabase)

1. Clone & Install

git clone <YOUR_GIT_URL>
cd <YOUR_PROJECT_NAME>
npm install

2. Start Local Supabase

This will spin up a local Postgres database and apply all migrations from supabase/migrations/:

supabase start

After starting, the CLI will output local credentials:

API URL:    http://127.0.0.1:54321
anon key:   eyJhbGci...
service_role key: eyJhbGci...
DB URL:     postgresql://postgres:postgres@127.0.0.1:54322/postgres

3. Configure Environment

Create a .env.local file in the project root (do not commit this file):

VITE_SUPABASE_URL=http://127.0.0.1:54321
VITE_SUPABASE_PUBLISHABLE_KEY=<anon_key_from_supabase_start>

Use .env.frontend.example for Vite (VITE_*) and .env.example for the Express backend (DATABASE_URL, JWT_SECRET, etc.); merge both into .env.local for local dev.

3.0 Admin login (Express + UI)

The React app and Express API are gated behind a single admin account (JWT in an httpOnly cookie).

1. Add to .env.local (same file as DATABASE_URL):

   JWT_SECRET=<output of: openssl rand -hex 32>

2. Apply DB migrations (includes app_admin_users):

   npm run db:migrate

3. Set a strong password and seed the admin user (default username admin):

   ADMIN_SEED_PASSWORD=your_secure_password_at_least_8_chars

   npm run db:seed:admin

4. Start the backend (npm run server) and frontend (npm run dev), then open /login.

Note: This protects the Express API and downloaded files under /files. The Supabase client in the browser still uses the public anon key; for stronger data isolation you would add Supabase Auth + RLS separately.

3.1 Connect Your Own Supabase Project (Cloud)

If you want to use your own hosted Supabase project instead of the default Lovable one:

1. Create a new project in the Supabase dashboard.

2. Link this repo to that project:

   supabase login
   supabase link --project-ref <your_project_ref>

3. Apply the schema + seed data to your project:

   supabase db reset

4. In .env.local, set your cloud project values:

   VITE_SUPABASE_URL=https://<your-project-ref>.supabase.co
   VITE_SUPABASE_PUBLISHABLE_KEY=<your_anon_public_key>

5. Set edge function secrets (service role key and external APIs) for that project:

   supabase secrets set SUPABASE_URL=https://<your-project-ref>.supabase.co
   supabase secrets set SUPABASE_SERVICE_ROLE_KEY=<your_service_role_key>
   supabase secrets set SCREENER_SESSION_ID=<your_screener_session_cookie>
   supabase secrets set SCREENER_CSRF_TOKEN=<your_screener_csrf_token>

6. Deploy the edge functions:

   supabase functions deploy --project-ref <your_project_ref> --all

4. Configure Edge Function Secrets

Edge functions need secrets to operate. Set them for local development:

# Required for edge functions to access the database
supabase secrets set SUPABASE_URL=http://127.0.0.1:54321
supabase secrets set SUPABASE_SERVICE_ROLE_KEY=<service_role_key_from_supabase_start>

# Required for Screener.in financial data scraping
supabase secrets set SCREENER_SESSION_ID=<your_screener_session_cookie>
supabase secrets set SCREENER_CSRF_TOKEN=<your_screener_csrf_token>

How to get Screener.in credentials:

1. Go to screener.in and log in
2. Open browser DevTools → Application → Cookies
3. Copy the values of sessionid and csrftoken

5. Serve Edge Functions Locally

In a separate terminal:

supabase functions serve

6. Start the Frontend

npm run dev

The app will be available at http://localhost:5173.

---

Project Structure

├── src/
│   ├── components/       # React components
│   ├── hooks/            # Custom React hooks (useStocks, useFinancials, etc.)
│   ├── integrations/     # Auto-generated Supabase client & types
│   ├── lib/              # Utilities, signals detection, types
│   └── pages/            # Route pages (Index, StocksPage, StockDetailPage, etc.)
├── supabase/
│   ├── config.toml       # Supabase project config
│   ├── migrations/       # Database migrations (auto-applied on `supabase start`)
│   └── functions/        # Edge functions
│       ├── fetch-price/          # Yahoo Finance price fetcher
│       ├── fetch-financials/     # Screener.in financial data scraper
│       ├── fetch-deals/          # Bulk/insider deal fetcher
│       ├── fetch-sector-indices/ # Nifty sector index tracker
│       ├── fetch-results-calendar/ # Upcoming results date fetcher
│       ├── refresh-all-prices/   # Batch price refresh
│       ├── refresh-all-financials/ # Batch financials refresh
│       └── analyze-transcript/   # AI transcript analysis

Key Features

• Auto-import on stock add: Price + financials fetched automatically
• 3Y price backfill: Historical daily prices from Yahoo Finance
• Earnings call analysis: Paste transcripts → AI extracts signals, promises, thesis drift
• Multibagger signal detection: Automated scoring based on financials, shareholding, management credibility
• Thesis tracking: Track investment thesis drift across quarters
• Sector indices: Nifty sector performance comparison
• Announcements: On each stock’s detail page, an Announcements tab lists downloaded filings (earnings, concall transcripts, investor presentations) by quarter and type with View links. Use Download filings to fetch from NSE & Screener.
• Google Drive upload (optional): Upload announcements to a Drive folder in a structured layout. See docs/GOOGLE_DRIVE_SETUP.md for creating a service account and API key.

Database

All tables have RLS enabled. Migrations in supabase/migrations/ are applied automatically when you run supabase start. Key tables:

Table	Purpose
stocks	Portfolio stocks with thesis & tracking config
prices	Daily price history
financial_metrics	Annual financial data (ROCE, ROE, etc.)
financial_results	Quarterly results
shareholding	Quarterly shareholding pattern
peer_comparison	Peer company metrics
transcript_analysis	AI-analyzed earnings call data
quarterly_snapshots	V5 quarterly thesis snapshots
management_promises	Tracked management commitments
bulk_deals / insider_trades	Deal activity
sector_indices	Nifty sector index prices

Useful Commands

supabase start          # Start local Supabase (applies migrations)
supabase stop           # Stop local Supabase
supabase db reset       # Reset DB and re-apply all migrations
supabase functions serve # Serve edge functions locally
npm run dev             # Start frontend dev server
npm run build           # Production build

Deployment

The app is deployed via Lovable. Click Publish in the editor to deploy.

Edge functions deploy automatically on code changes. Frontend requires clicking Update in the publish dialog.