ChatbotUI

Full-stack financial chatbot with daily briefings, watchlist management, RAG-based Q&A, and on-demand ticker summarization.

Architecture

ChatbotUI/
├── backend/          # FastAPI BFF + briefing scheduler
│   ├── main.py       # API server (watchlist, briefings, chat, summarize)
│   ├── briefing.py   # Scheduled daily briefing generator
│   ├── rag.py        # RAG pipeline (embed query → vector search → LLM)
│   ├── fetcher.py    # Data extraction from Supabase
│   ├── summarizer.py # Moonshot API for summaries & briefings
│   ├── db.py         # Supabase data access layer
│   └── config.py     # Environment configuration
├── frontend/         # React + Vite
│   └── src/
│       ├── pages/
│       │   ├── BriefingPage.jsx   # View daily briefings
│       │   ├── WatchlistPage.jsx  # Manage tracked tickers + preferences
│       │   └── ChatPage.jsx       # RAG chat + on-demand summarize
│       ├── components/
│       │   └── Sidebar.jsx
│       ├── api.js    # API client
│       └── App.jsx
└── README.md

Prerequisites

• Python 3.11+
• Node.js 18+
• Running embedding-service (port 8002)
• Supabase project with schema applied (see data-pipeline/pipeline/schema.sql)
• Moonshot API key

Quick Start

Backend

cd ChatbotUI/backend
cp .env.example .env     # fill in credentials
pip install -r requirements.txt
uvicorn main:app --port 8000 --reload

Frontend

cd ChatbotUI/frontend
npm install
npm run dev              # http://localhost:3000

Daily Briefing (manual test)

cd ChatbotUI/backend
python briefing.py

In production, briefing.py runs hourly via GitHub Actions (.github/workflows/daily-briefing.yml). It checks each user's timezone and generates a briefing if their local time is 08:00-08:14.

API Endpoints

Method	Path	Description
GET	/api/tickers	Available tickers for watchlist
GET	/api/watchlist	User's watchlist
POST	/api/watchlist	Add ticker to watchlist
DELETE	/api/watchlist/{ticker}	Remove ticker
GET	/api/preferences	User preferences (timezone, briefing toggle)
PUT	/api/preferences	Update preferences
GET	/api/briefings	Recent daily briefings
GET	/api/briefings/latest	Most recent briefing
POST	/api/chat	RAG-based Q&A
POST	/api/summarize	On-demand ticker summary
GET	/health	Health check

Integration with Other Services

• data-pipeline — Populates Supabase with documents, earnings, price data
• embedding-service — Encodes queries for RAG vector search (port 8002)
• Summarization — Standalone CLI summarization (ChatbotUI backend reuses the same logic)
• Supabase — Central database for all data + user state