Documentation

Everything you need to know about the Trading Agent System — from getting started to advanced features.

Getting Started

Trading Agent System is an autonomous trading platform that combines AI-powered research, quantitative analysis, and visual chart judgment to help you make informed trading decisions. It supports the full workflow: research → strategy → backtest → paper trade → live execution.

Quick Start Tutorial

Complete Onboarding

Set your experience level, risk tolerance, and investment goals in the onboarding wizard.

Run Market Research

Go to Market Research and run a daily analysis to understand current market conditions.

Design a Strategy

Use Strategy Chat to interactively build a trading strategy with the AI assistant.

Backtest It

Head to Backtesting, select your strategy and symbols, set a date range, and run a historical test.

Paper Trade

Start paper trading to test your strategy in real-time with virtual capital before going live.

Feature Reference

Market Research

AI-powered macro-economic analysis using Perplexity AI. Get daily market overviews, sector trends, and economic indicators.

Stock Research

Deep fundamental analysis on individual stocks. AI agents evaluate financials, competitive position, and growth prospects.

SEC Filings

Browse and analyze 10-K, 10-Q, and 8-K filings from SEC EDGAR with AI-powered insights.

Sentiment Analysis

Real-time news sentiment scoring. Aggregate sentiment by symbol or view overall market mood.

Academic Research

Search arXiv for finance and trading papers. AI analyzes papers for practical trading applicability.

Strategy Chat

Interactive AI conversations to design custom trading strategies. The AI helps define entry/exit rules, risk parameters, and more.

Charts

OHLCV candlestick data across multiple timeframes (1m to monthly) with intelligent caching.

Backtesting

Test strategies against historical data. Configure capital, commission, slippage, and position sizing.

Paper Trading

Simulated trading with $100K virtual capital. Place orders, track positions, and measure performance risk-free.

Broker Integration

Connect to Alpaca for live trading. Place real orders, monitor positions, and manage your brokerage account.

Risk Management

Configure position limits, stop-loss rules, and max drawdown thresholds. Apply preset risk profiles.

Analytics

Performance analytics including win rate, Sharpe ratio, max drawdown, and equity curves from your trading history.

Execution Agent

AI-powered buy/sell recommendations combining chart analysis, research, and strategy rules.

API Reference

82 RESTful endpoints organized across 23 service areas. All endpoints require Firebase Authentication via Bearer token in the Authorization header.

Interactive API Explorer

Dark-themed Swagger UI embedded in the app. Try endpoints directly.

Swagger UI (Native)

FastAPI's built-in Swagger interface at the backend URL.

ReDoc

Clean, readable three-panel API documentation.

Architecture

Backend (Python)

• Framework: FastAPI with async support
• AI Agents: Pydantic AI with Claude & Perplexity
• Data: yfinance, SEC EDGAR, arXiv, Google News RSS
• Auth: Firebase Authentication (JWT)
• Storage: Firebase Firestore (per-user isolation)
• Broker: Alpaca API (paper + live trading)
• Package Manager: uv (fast Python packaging)

Frontend (TypeScript)

• Framework: Next.js 15 (App Router)
• UI: shadcn/ui + Tailwind CSS
• Charts: Recharts (equity curves, distributions)
• Streaming: SSE for real-time chat responses
• Auth: Firebase client SDK
• Responsive: Mobile-first with hamburger navigation
• Theme: Dark mode with emerald accent palette

Agent Pipeline

Market Research→Stock Research→SEC Analysis→Sentiment Scoring→Strategy Evaluation→Risk Check→Execution Decision→Order + Reasoning Log

Every execution decision creates a reasoning log documenting which data sources contributed, their impact (bullish/bearish/neutral), confidence levels, and risk assessment.