python-telegrambot/ai-auto-trading — explained in plain English
Analysis updated 2026-05-18
Test an AI-driven trading strategy safely on Gate.io or Binance testnets before risking real money.
Let a large language model analyze markets and decide trade entries, exits, and position sizing automatically.
Use dynamic ATR-based stop-losses and partial profit taking to manage risk on open positions.
Monitor the bot's live status and trade history through a health dashboard and local SQLite logs.
| python-telegrambot/ai-auto-trading | tsukamg/polymarket-weather-trading-engine | sinotrade/shioaji-pro-app | |
|---|---|---|---|
| Stars | 163 | 168 | 157 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | hard | moderate | hard |
| Complexity | 5/5 | 4/5 | 4/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires exchange API keys, an AI model API, and careful testnet validation before any live trading.
NexusQuant is an AI-powered cryptocurrency trading bot written in TypeScript that connects to crypto exchanges and executes trades automatically based on market analysis performed by a large language model. The bot treats the AI model as the core decision-maker, it reads market data, selects trading strategies, manages open positions, and controls risk without requiring manual input. It connects to Gate.io and Binance for perpetual futures trading (contracts that let you bet on whether a crypto asset will go up or down). Before trading live, both exchanges offer testnets (simulated environments with fake money) where you can verify the bot's behavior safely. The risk management system is a notable focus: stop-losses are calculated dynamically based on recent volatility (using the ATR, Average True Range, metric), positions are only partially closed at each profit target rather than all at once, and stop-losses are automatically moved to break-even once a trade is sufficiently profitable. The bot also scores each potential trade across four factors before entering, and only acts on trades that meet a minimum threshold. A health monitoring dashboard tracks the system's status in real time. Multiple AI models are supported through a compatible API, including options from various providers. The bot logs all decisions, trades, and account history to a local SQLite database. Node.js 20.19 or newer is required. The full README is longer than what was provided.
An AI-driven crypto trading bot that lets a language model analyze markets and automatically execute perpetual futures trades on Gate.io and Binance.
Mainly TypeScript. The stack also includes TypeScript, Node.js, SQLite.
You can use and modify this freely, but if you run a modified version as a network service, you must offer its source code to users of that service (AGPL-3.0).
Setup difficulty is rated hard, with roughly 1day+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.