🎓 NOFX+: Real-time Feedback AI Trading Platform - Developer Onboarding Guide

⚠️ Important: This repository contains NOFX+ (nofxplus) - the production-hardened fork with enhancements.

• Local Setup (./setup.sh): Runs NOFX+ (your version) with all enhancements
• Docker Deployment (install.sh): Pulls original NOFX from official repo (no nofxplus enhancements)

For NOFX+ features, use local setup. Docker is for official NOFX deployment.

🚀 From NOFX to NOFX+ - The Production Evolution

NOFX+ (nofxplus) is a production-hardened fork of the original NOFX trading system that:

• Fixes 17 critical issues See Merge Request Logs
• Adds LLM-evolve feedback + prompt variants, dynamic threshold calibration, failure analysis, and microstructure intelligence
• Delivers institutional-grade trading performance

Why NOFX+? While NOFX pioneered LLM-driven trading, NOFX+ adds LLM-evolve feedback loops, prompt variant evolution, market microstructure intelligence, and adaptive threshold calibration missing from the original implementation.

Welcome! This guide will take you from zero to complete mastery of the NOFX+ codebase.

⭐ If you find NOFX+ useful, please give the repo a star.

---

📖 Table of Contents

1. Quick Start - Your First 30 Minutes
2. System Architecture Overview
3. System Directory Walkthrough
4. NoFx Feedback Mechanism Benchmark
5. Core Components Deep Dive
6. Trade Failure Analysis & Feedback Loop
7. Integration & Usage Verification
8. Code Audit & Unused Functions
9. Getting Started
10. Contributing
11. NoFx+ Screenshots
12. NoFx Original Repo

---

1. Quick Start - Your First 30 Minutes

Read These Files First (in order)

# 1. System overview (5 min)
docs/README.md              # Documentation index
README.md                   # Project overview

# 2. Architecture understanding (10 min)
docs/architecture/README.md # System architecture
main.go                     # Application entry point

# 3. Configuration (5 min)
config/config.go           # Global config structure
.env.example               # Environment variables

# 4. Core flow (10 min)
manager/trader_manager.go  # Trader orchestration
decision/engine.go         # AI decision making (lines 1-200)

---

2. System Architecture Overview

Evolution: NOFX → NOFX+

NOFX (Original):

[Market Data Polling] → [LLM Decision] → [Basic Execution] → [Simple P&L Tracking]

NOFX+ (Enhanced):

┌─────────────────────────────────────────────────────────────────────────┐
│                       REAL-TIME DATA LAYER                              │
│  [WebSocket Streams] → [Order Book Monitor] → [Market Microstructure]   │
└─────────────────────────────────────────────────────────────────────────┘
                                    ↓
┌─────────────────────────────────────────────────────────────────────────┐
│                    INTELLIGENT DECISION ENGINE                          │
│  [Configurable Indicators] → [LLM + Microstructure] → [Risk Validation] │
└─────────────────────────────────────────────────────────────────────────┘
                                    ↓
┌─────────────────────────────────────────────────────────────────────────┐
│                  EXECUTION & LEARNING LOOP                              │
│  [Smart Execution] → [LLM-Enhanced Feedback] → [Prompt Evolution]       │
│         ↑                    ↓                    ↓                     │
│  [Compliance Tracking] ← [Failure Analysis] ← [Threshold Calibration]   │
└─────────────────────────────────────────────────────────────────────────┘

Data-driven failure analysis replaces magic numbers: The feedback system automatically analyzes your live trading performance and provides:

1. Real-time Performance Metrics - Win rate, profit factor, Sharpe ratio, drawdown
2. Pattern Recognition - Success and failure patterns identified from recent trades
3. AI-Generated Insights - LLM-powered analysis of what's working and what needs improvement
4. Actionable Recommendations - Specific rules and adjustments based on performance data
5. Prompt Evolution - Continuously optimizes the trading prompt based on feedback
6. Dynamic Threshold Calibration - Uses trade outcome metrics to calibrate decision thresholds based on market microstructure and performance patterns

---

3. NOFX+ System Directory Walkthrough

nofx/
├── main.go                    # Entry point - START HERE
├── config/                    # Global configuration
│   └── config.go             # Config struct + initialization
├── manager/                   # Trader orchestration
│   ├── trader_manager.go     # Multi-trader coordination
│   └── trader_manager_test.go
├── trader/                    # Trading execution engine
│   ├── trader.go             # Core trader loop 🚀 OPTIMIZED
│   ├── trader_*.go           # Exchange-specific implementations 🚀 OPTIMIZED
│   └── trader_papertrading.go # Paper trading mode
├── decision/                  # AI decision making ⭐ KEY MODULE
│   ├── engine.go             # Decision orchestration 🚀 OPTIMIZED
│   ├── schema.go             # AI system prompt construction 🔥 NEW
│   ├── formatter.go          # AI user prompt construction 🔥 NEW
│   ├── trade_failure.go      # Failure analysis 🔥 NEW
│   └── threshold_calibrator.go # Data-driven thresholds 🔥 NEW
├── market/                    # Market data & microstructure ⭐ KEY MODULE
│   ├── api_client.go         # Exchange API client
│   ├── data.go               # Market data aggregation 🚀 OPTIMIZED
│   ├── microstructure.go     # Order book analysis 🔥 NEW
│   ├── timeframe.go          # Multi-timeframe logic
│   ├── *_websocket.go        # Real-time data streams
│   └── order_book_monitor.go # Liquidity monitoring 🚀 OPTIMIZED
├── backtest/                  # Backtesting engine ⭐ KEY MODULE
│   ├── manager.go            # Backtest orchestration 🚀 OPTIMIZED
│   ├── runner.go             # Simulation execution 🚀 OPTIMIZED
│   ├── feedback.go           # LLM-evolve feedback system 🔥 NEW
│   ├── prompt_optimizer.go   # Prompt variant evolution 🔥 NEW
│   ├── factor_optimizer.go   # Risk control optimization 🔥 NEW
│   ├── compliance_tracker.go # Reinforcement compliance tracking 🔥 NEW
│   ├── smart_heuristics.go   # Adaptive position sizing 🔥 NEW
│   ├── calibration.go        # Threshold calibration pipeline 🔥 NEW
│   ├── account.go            # Position & PnL tracking 🚀 OPTIMIZED
│   ├── metrics.go            # Performance metrics 🚀 OPTIMIZED
│   └── persistence_db.go     # Results storage
├── store/                     # Database layer
│   ├── store.go              # Main store interface 🚀 OPTIMIZED
│   ├── trader.go             # Trader persistence 🚀 OPTIMIZED
│   ├── position.go           # Position tracking 🚀 OPTIMIZED
│   └── position_builder.go   # Position lifecycle
│   └── trade_outcome.go       # Trade outcome metrics 🔥 NEW
├── api/                       # REST API server
│   ├── server.go             # API routes and handlers 🚀 OPTIMIZED
│   ├── strategy.go           # Strategy endpoints 🚀 OPTIMIZED
│   ├── backtest.go           # Backtest endpoints 🚀 OPTIMIZED
│   └── debate.go             # Debate arena endpoints 🚀 OPTIMIZED
├── mcp/                       # AI provider clients
│   ├── claude_client.go      # Anthropic Claude
│   ├── deepseek_client.go    # DeepSeek
│   └── openai_client.go      # OpenAI/compatible
├── debate/                    # Multi-AI debate system
│   └── engine.go             # Debate orchestration 🚀 OPTIMIZED
├── web/                       # Frontend (React/TypeScript)
│   ├── src/
│   │   ├── App.tsx           # Main app component
│   │   ├── components/       # UI components 🚀 OPTIMIZED
│   │   ├── lib/              # API client
│   │   └── stores/           # State management 🚀 OPTIMIZED
│   └── package.json
└── docs/                      # Documentation
    ├── architecture/          # Architecture docs
    ├── getting-started/       # Deployment guides
    ├── guides/                # User guides

---

4. NOFX+ Learning Systems Benchmark

NOFX+: AI Trading That Actually Learns

📊 The Performance

Before NOFX+ (Naive LLM trading):

• 📉 -27.9% total return
• 📉 34.8% win rate (essentially random)
• 📉 -0.03 Sharpe Ratio (negative risk-adjusted returns)
• 📉 0.19 Profit Factor (losing $5 for every $1 made)
• 📉 27.9% max drawdown

With NOFX+ Feedback (Enabled at cycle 156):

• 📈 +11.6% total return (+39.5% improvement)
• 📈 66.7% win rate (+91% improvement)
• 📈 3.35 Profit Factor (making $3.35 for every $1 lost)
• 📈 4.9% max drawdown (82% reduction)
• 📈 ETHUSDT: 100% win rate (3/3 trades)

With NOFX+ LLM-Evolve Feedback + System Prompt Evolution (Enabled at cycle 34):

• 📈 +12.5% total return (+7.6% improvement than Feedback Analysis only)
• 📈 61.5% win rate
• 📈 6.13 Profit Factor (+183% improvement than Feedback Analysis only)
• 📈 4.3% max drawdown (12% reduction than Feedback Analysis only)
• 📈 BNBUSDT & DOGEUSDT: 100% win rate (5/5 trades)

The NOFX+ Learning Stack (What Changed)

We introduced a LLM-evolve learning stack that:

1. Analyzes every trade with microstructure-aware evidence
2. Explains failures via deterministic failure analysis
3. Calibrates thresholds dynamically (no more magic numbers)
4. Evolves prompt variants to improve decision quality over time
5. Feeds calibrated thresholds back into the decision context
6. Optimizes risk controls via the factor optimizer (inner-loop tuning)

The Results

{
  "improvement": {
    "total_return": "+40.4%",
    "win_rate": "+91%",
    "profit_factor": "+3,263%",
    "max_drawdown": "-85%",
    "avg_win_size": "+150%",
    "avg_loss_size": "-20%"
  }
}

5. Core Components Deep Dive

5.1 Application Startup (main.go)

What happens when you start NOFX:

1. Configuration loading from .env
2. Database initialization
3. Market data connection setup
4. Trader manager instantiation
5. API server startup

Key files to understand startup:

• main.go (Lines 1-170)
• config/config.go (Lines 1-120)

5.2 Trading Loop (trader/auto_trader.go)

Read these files in order:

1. trader/auto_trader.go - Core live trader (Lines 1-300)
2. backtest/runner.go - Backtest trade runner (Lines 1-400)
3. decision/engine.go - AI decision (Lines 1-500)
4. market/data.go - Market data (Lines 1-200)

5.3 Decision Engine (decision/engine.go)

Critical files for understanding AI decisions:

1. decision/engine.go - Main orchestration
2. decision/formatter.go - Prompt construction
3. decision/schema.go - Orchestrate rules
4. backtest/feedback.go - Feedback prompt construction
5. backtest/compliance_tracker.go - Track adherence to recommendations
6. backtest/prompt_optimizer.go - Optimize prompt variants
7. backtest/factor_optimizer.go - Optimize risk control factors
8. backtest/smart_heuristics.go - Adaptive heuristics for leverage control
9. decision/threshold_calibrator.go - Data-driven threshold calibration

5.4 Backtest Engine (backtest/)

Files to understand:

1. backtest/manager.go - Orchestration
2. backtest/runner.go - Execution engine (optimized)
3. backtest/account.go - Position tracking
4. backtest/metrics.go - Performance metrics (optimized)

5.5 Live Market Data System (market/)

Files to understand market data:

1. market/data.go - Data aggregation (optimized)
2. market/microstructure.go - Order book analysis (optimized)
3. market/timeframe.go - Multi-timeframe logic
4. market/binance_websocket.go - Real-time streams

---

6. Trade Failure Analysis & Feedback Loop

System Overview

Trading Execution
    ↓
Trader.DoCycle()
    ↓
Calculate Stats & Historical Trades (20+)
    ↓
Check: TotalTrades >= MinDecisionsForFeedback?
    ↓ YES
Analyze winning/losing patterns
    ↓
FeedbackGenerator.Generate(LLM)Feedback
    ↓
Return FeedbackAnalysis & Actionable Recommendations
    ↓
Store in lastFeedback & save to disk
    ↓
PromptOptimizer uses feedback to evolve prompts
        ↓
   Trade Outcome Metrics
   (volume, OI, spread, depth)
        ↓
   ROC Analysis + Youden's J
   (Find optimal thresholds)
        ↓
   Calibrated Thresholds
    ↓
Compliance Tracking
    ↓
Next decision uses evolved prompt

---

7. Integration & Usage Verification

✅ All Systems Verified Working

Getting Started:

✅ Follow the Getting Started guide to deploy NOFX+ on your local machine or server.

Build Status:

✅ go build ./...           # All packages compile
✅ go vet ./...             # Zero linter warnings
✅ go test ./backtest       # All tests pass
✅ go test ./decision       # All tests pass
✅ npm run build (web/)     # Frontend builds
✅ make build               # Backend builds
✅ make build-frontend      # Frontend builds
✅ make deps-update         # Dependencies up to date
✅ make deps-frontend       # Frontend deps up to date
✅ make fmt                 # Code formatted
✅ make run-frontend        # Frontend runs
✅ make run                 # Backend runs

---

8. Code Audit & Unused Functions

8.1 How to Find Unused Code

Method 1: Use make lint

# Run linter to find unused code
make lint
# Check output for "unused" warnings

8.2 Complete Audit Script

Method 1: Use make lint

# Run linter to find unused code
make lint
# Check output for "unused" warnings
# Check output for "staticcheck" warnings
# Check output for "errcheck" warnings

Method 2: Use make fmt

# Run formatter to format code
make fmt
# Check for any formatting issues

Method 3: Custom Audit Script

#!/bin/bash
# audit_unused.sh

echo "🔍 NOFX Code Audit - Finding Unused Functions"
echo "=============================================="

# Find all function definitions
echo "📝 Scanning all functions..."
find . -name "*.go" -exec grep -Hn "^func " {} \; | \
    grep -v "_test.go" | \
    grep -v "vendor/" > /tmp/all_funcs.txt

total=$(wc -l < /tmp/all_funcs.txt)
echo "Found $total functions"

# ... (rest of audit logic)

---

9. Getting Started

✅ Data Source Update: The data pooling API has been updated from http://nofxaios.com:30006/api to https://nofxos.ai. NOFX+ implements a Binance + Coinglass + nofxos.ai hybrid method for reliable market data, Open Interest (OI), and quantitative metrics. Note: This repository's API integration updates may not be actively maintained in the short term as development focus shifts to other projects.

💡 Recommended Workflow: Run a backtest first before starting live trading. Live trading will load strategy parameters from your backtest results if available. Starting without a backtest means creating a fresh strategy from scratch, requiring ~1 week of live trading data before the feedback mechanism can optimize your strategy, and even longer for prompt evolution to take effect. This is because we need sufficient behavioral data to statistically optimize your strategy and decision thresholds.

Quick Start (Automated - 5 Minutes)

The easiest way - let our setup script handle everything:

# 1. Clone repository
git clone https://github.com/jeffeehsiung/nofxplus.git
cd nofxplus

# 2. Run automated setup (macOS/Linux)
./setup.sh

# 3. Start the application
make run                 # Backend
make run-frontend        # Frontend (in another terminal)

# 4. Open in browser
http://localhost:3000

That's it! The setup script will:

• ✅ Check all prerequisites (Go, Node.js, TA-Lib)
• ✅ Install dependencies
• ✅ Generate encryption keys
• ✅ Build backend & frontend
• ✅ Verify everything is working (27 checks)
• ✅ Show startup instructions

Manual Setup (Windows / Advanced Users)

For detailed setup instructions, multiple OS options, Docker deployment, and troubleshooting:

👉 See SETUP_GUIDE.md for complete documentation

Access Your NOFX+ Instance

Once running, open your browser:

http://localhost:3000

Then:

1. Configure AI Models - Add your AI API keys (DeepSeek, OpenAI, Claude, etc.)
2. Configure Exchanges - Set up Binance, Bybit, or other exchange credentials
3. Create Strategy - Build your first trading strategy in Strategy Studio
4. Create Trader - Combine AI model + Exchange + Strategy
5. Start Trading - Launch your first trader!

---

10. Contributing

Areas Needing Improvement

1. Architecture refactoring for modularity
2. Enhanced code quality, design pattern, and documentation
3. More comprehensive testing
4. More exchange integrations (Kraken, Coinbase, etc.)
5. Additional microstructure indicators
6. Advanced machine learning models

Core Team

NOFX+

• Jeffee Hsiung - jeffeehsiung - Lead Developer & Architect NOFX+ is built upon the groundbreaking work of the NOFX team, enhanced with production hardening, market microstructure intelligence, and adaptive learning algorithms developed through extensive backtesting and real trading experience.

NOFX Team

• Original NOFX development team (see NOFX GitHub)
• Tinkle - @Web3Tinkle
• Official Twitter - @nofx_official

Contribution Guidelines

# 1. Fork repository
# 2. Create feature branch
git checkout -b feature/your-feature
# 3. Make changes with tests
# 4. Run verification
# 5. Submit pull request

Code Standards

• Go: gofmt, go vet, 80%+ test coverage
• Frontend: TypeScript, ESLint, Prettier
• Documentation: Update README and relevant docs
• Testing: Include unit and integration tests

---

11. NoFx+ Screenshots

Live Traders w/ Prompt Variant Evolution & Feedback Analysis Display

Live Prompt Variants & Feedback	Concised Feedback Embedded in UserPrompt

Backtest

Backtest Feedback	Backtest Prompt Lab

Backtest Result + Feedback + Prompt Variants

---

📄 License

MIT License - See LICENSE file for details.

Built upon the groundbreaking work of the NOFX team, enhanced with production hardening, market microstructure intelligence, and adaptive learning algorithms developed through extensive backtesting and real trading experience.

---

Last Updated: January 2026 | Version: NOFX+ 1.0.0 | Contributors: NOFX Community + Jeffee Enhancements

---

12. NOFX Original Repo

🚀 Original NOFX (shout out to the team!)

---

NOFX - Agentic Trading OS

CONTRIBUTOR AIRDROP PROGRAM
Code · Bug Fixes · Issues → Airdrop
Learn More

Languages: English | 中文 | 日本語 | 한국어 | Русский | Українська | Tiếng Việt

---

AI-Powered Multi-Asset Trading Platform

NOFX is an open-source AI trading system that lets you run multiple AI models to trade automatically. Configure strategies through a web interface, monitor performance in real-time, and let AI agents compete to find the best trading approach.

Supported Markets

Market	Trading	Status
🪙 Crypto	BTC, ETH, Altcoins	✅ Supported
📈 US Stocks	AAPL, TSLA, NVDA, etc.	✅ Supported
💱 Forex	EUR/USD, GBP/USD, etc.	✅ Supported
🥇 Metals	Gold, Silver	✅ Supported

Core Features

• Multi-AI Support: Run DeepSeek, Qwen, GPT, Claude, Gemini, Grok, Kimi - switch models anytime
• Multi-Exchange: Trade on Binance, Bybit, OKX, Bitget, Hyperliquid, Aster DEX, Lighter from one platform
• Strategy Studio: Visual strategy builder with coin sources, indicators, and risk controls
• AI Debate Arena: Multiple AI models debate trading decisions with different roles (Bull, Bear, Analyst)
• AI Competition Mode: Multiple AI traders compete in real-time, track performance side by side
• Web-Based Config: No JSON editing - configure everything through the web interface
• Real-Time Dashboard: Live positions, P/L tracking, AI decision logs with Chain of Thought

Core Team

• Tinkle - @Web3Tinkle
• Official Twitter - @nofx_official

Risk Warning: This system is experimental. AI auto-trading carries significant risks. Strongly recommended for learning/research purposes or testing with small amounts only!

Developer Community

Join our Telegram developer community: NOFX Developer Community

---

Screenshots

Config Page

AI Models & Exchanges	Traders List

Competition & Backtest

Competition Mode	Backtest Lab

Dashboard

Overview	Market Chart

Trading Stats	Position History

Positions	Trader Details

Strategy Studio

Strategy Editor	Indicators Config

Debate Arena

AI Debate Session	Create Debate

---

Supported Exchanges

CEX (Centralized Exchanges)

Exchange	Status	Register (Fee Discount)
Binance	✅ Supported	Register
Bybit	✅ Supported	Register
OKX	✅ Supported	Register
Bitget	✅ Supported	Register

Perp-DEX (Decentralized Perpetual Exchanges)

Exchange	Status	Register (Fee Discount)
Hyperliquid	✅ Supported	Register
Aster DEX	✅ Supported	Register
Lighter	✅ Supported	Register

---

Supported AI Models

AI Model	Status	Get API Key
DeepSeek	✅ Supported	Get API Key
Qwen	✅ Supported	Get API Key
OpenAI (GPT)	✅ Supported	Get API Key
Claude	✅ Supported	Get API Key
Gemini	✅ Supported	Get API Key
Grok	✅ Supported	Get API Key
Kimi	✅ Supported	Get API Key

---

Quick Start

One-Click Install (Recommended)

Linux / macOS:

curl -fsSL https://raw.githubusercontent.com/NoFxAiOS/nofx/main/install.sh | bash

That's it! Open http://127.0.0.1:3000 in your browser.

Docker Compose (Manual)

# Download and start
curl -O https://raw.githubusercontent.com/NoFxAiOS/nofx/main/docker-compose.prod.yml
docker compose -f docker-compose.prod.yml up -d

Access Web Interface: http://127.0.0.1:3000

# Management commands
docker compose -f docker-compose.prod.yml logs -f    # View logs
docker compose -f docker-compose.prod.yml restart    # Restart
docker compose -f docker-compose.prod.yml down       # Stop
docker compose -f docker-compose.prod.yml pull && docker compose -f docker-compose.prod.yml up -d  # Update

Keeping Updated

💡 Updates are frequent. Run this command daily to stay current with the latest features and fixes:

curl -fsSL https://raw.githubusercontent.com/NoFxAiOS/nofx/main/install.sh | bash

This one-liner pulls the latest official images and restarts services automatically.

Manual Installation (For Developers)

Prerequisites

• Go 1.21+
• Node.js 18+
• TA-Lib (technical indicator library)

# Install TA-Lib
# macOS
brew install ta-lib

# Ubuntu/Debian
sudo apt-get install libta-lib0-dev

Installation Steps

# 1. Clone the repository
git clone https://github.com/NoFxAiOS/nofx.git
cd nofx

# 2. Install backend dependencies
go mod download

# 3. Install frontend dependencies
cd web
npm install
cd ..

# 4. Build and start backend
go build -o nofx
./nofx

# 5. Start frontend (new terminal)
cd web
npm run dev

Access Web Interface: http://127.0.0.1:3000

---

Windows Installation

Method 1: Docker Desktop (Recommended)

1. Install Docker Desktop

   • Download from docker.com/products/docker-desktop
   • Run the installer and restart your computer
   • Start Docker Desktop and wait for it to be ready

2. Run NOFX

   # Open PowerShell and run:
   curl -o docker-compose.prod.yml https://raw.githubusercontent.com/NoFxAiOS/nofx/main/docker-compose.prod.yml
   docker compose -f docker-compose.prod.yml up -d

3. Access: Open http://127.0.0.1:3000 in your browser

Method 2: WSL2 (For Development)

1. Install WSL2

   # Open PowerShell as Administrator
   wsl --install

   Restart your computer after installation.

2. Install Ubuntu from Microsoft Store

   • Open Microsoft Store
   • Search "Ubuntu 22.04" and install
   • Launch Ubuntu and set up username/password

3. Install Dependencies in WSL2

   # Update system
   sudo apt update && sudo apt upgrade -y
   
   # Install Go
   wget https://go.dev/dl/go1.21.5.linux-amd64.tar.gz
   sudo tar -C /usr/local -xzf go1.21.5.linux-amd64.tar.gz
   echo 'export PATH=$PATH:/usr/local/go/bin' >> ~/.bashrc
   source ~/.bashrc
   
   # Install Node.js
   curl -fsSL https://deb.nodesource.com/setup_18.x | sudo -E bash -
   sudo apt-get install -y nodejs
   
   # Install TA-Lib
   sudo apt-get install -y libta-lib0-dev
   
   # Install Git
   sudo apt-get install -y git

4. Clone and Run NOFX

   git clone https://github.com/NoFxAiOS/nofx.git
   cd nofx
   
   # Build and run backend
   go build -o nofx && ./nofx
   
   # In another terminal, run frontend
   cd web && npm install && npm run dev

5. Access: Open http://127.0.0.1:3000 in Windows browser

Method 3: Docker in WSL2 (Best of Both Worlds)

1. Install Docker Desktop with WSL2 backend

   • During Docker Desktop installation, enable "Use WSL 2 based engine"
   • In Docker Desktop Settings → Resources → WSL Integration, enable your Linux distro

2. Run from WSL2 terminal

   curl -fsSL https://raw.githubusercontent.com/NoFxAiOS/nofx/main/install.sh | bash

---

Server Deployment

Quick Deploy (HTTP via IP)

By default, transport encryption is disabled, allowing you to access NOFX via IP address without HTTPS:

# Deploy to your server
curl -fsSL https://raw.githubusercontent.com/NoFxAiOS/nofx/main/install.sh | bash

Access via http://YOUR_SERVER_IP:3000 - works immediately.

Enhanced Security (HTTPS)

For enhanced security, enable transport encryption in .env:

TRANSPORT_ENCRYPTION=true

When enabled, browser uses Web Crypto API to encrypt API keys before transmission. This requires:

• https:// - Any domain with SSL
• http://localhost - Local development

Quick HTTPS Setup with Cloudflare

1. Add your domain to Cloudflare (free plan works)

   • Go to dash.cloudflare.com
   • Add your domain and update nameservers

2. Create DNS record

   • Type: A
   • Name: nofx (or your subdomain)
   • Content: Your server IP
   • Proxy status: Proxied (orange cloud)

3. Configure SSL/TLS

   • Go to SSL/TLS settings
   • Set encryption mode to Flexible

   User ──[HTTPS]──→ Cloudflare ──[HTTP]──→ Your Server:3000
   

4. Enable transport encryption

   # Edit .env and set
   TRANSPORT_ENCRYPTION=true

5. Done! Access via https://nofx.yourdomain.com

---

Initial Setup (Web Interface)

After starting the system, configure through the web interface:

1. Configure AI Models - Add your AI API keys (DeepSeek, OpenAI, etc.)
2. Configure Exchanges - Set up exchange API credentials
3. Create Strategy - Configure trading strategy in Strategy Studio
4. Create Trader - Combine AI model + Exchange + Strategy
5. Start Trading - Launch your configured traders

All configuration is done through the web interface - no JSON file editing required.

---

Web Interface Features

Competition Page

• Real-time ROI leaderboard
• Multi-AI performance comparison charts
• Live P/L tracking and rankings

Dashboard

• TradingView-style candlestick charts
• Real-time position management
• AI decision logs with Chain of Thought reasoning
• Equity curve tracking

Strategy Studio

• Coin source configuration (Static list, AI500 pool, OI Top)
• Technical indicators (EMA, MACD, RSI, ATR, Volume, OI, Funding Rate)
• Risk control settings (leverage, position limits, margin usage)
• AI test with real-time prompt preview

Debate Arena

• Multi-AI debate sessions for trading decisions
• Configurable AI roles (Bull, Bear, Analyst, Contrarian, Risk Manager)
• Multiple rounds of debate with consensus voting
• Auto-execute consensus trades

Backtest Lab

• 3-step wizard configuration (Model → Parameters → Confirm)
• Real-time progress visualization with animated ring
• Equity curve chart with trade markers
• Trade timeline with card-style display
• Performance metrics (Return, Max DD, Sharpe, Win Rate)
• AI decision trail with Chain of Thought

---

Common Issues

TA-Lib not found

# macOS
brew install ta-lib

# Ubuntu
sudo apt-get install libta-lib0-dev

AI API timeout

• Check if API key is correct
• Check network connection
• System timeout is 120 seconds

Frontend can't connect to backend

• Ensure backend is running on http://localhost:8080
• Check if port is occupied

---

Documentation

Document	Description
Architecture Overview	System design and module index
Strategy Module	Coin selection, data assembly, AI prompts, execution
Backtest Module	Historical simulation, metrics, checkpoint/resume
Debate Module	Multi-AI debate, voting consensus, auto-execution
FAQ	Frequently asked questions
Getting Started	Deployment guide

---

License

This project is licensed under GNU Affero General Public License v3.0 (AGPL-3.0) - See LICENSE file.

---

Contributing

We welcome contributions! See:

• Contributing Guide - Development workflow and PR process
• Code of Conduct - Community guidelines
• Security Policy - Report vulnerabilities

---

Contributor Airdrop Program

All contributions are tracked on GitHub. When NOFX generates revenue, contributors will receive airdrops based on their contributions.

PRs that resolve Pinned Issues receive the HIGHEST rewards!

Contribution Type	Weight
Pinned Issue PRs	⭐⭐⭐⭐⭐⭐
Code Commits (Merged PRs)	⭐⭐⭐⭐⭐
Bug Fixes	⭐⭐⭐⭐
Feature Suggestions	⭐⭐⭐
Bug Reports	⭐⭐
Documentation	⭐⭐

---

Contact

• GitHub Issues: Submit an Issue
• Developer Community: Telegram Group

---

Star History

---