Skip to content

aipushmsg/binance-Quant-Zero

BranchesTags
 
 

Repository files navigation

🌌 领哥量化机甲 (Quant-Zero V1.0 满配版)

币安 AI 智能交易信号与 SMC 风控 Agent

基于 OpenClaw 的 AI 驱动交易 Agent，专为币安客户及量化极客打造。 本项目集成了大语言模型语义解析、SMC 机构级盘面推演与彭博终端级 UI 面板，彻底颠覆传统量化机器人的交互体验。

🎯 核心黑科技 (Core Features)

  • 🧠 SMC 机构级参谋大脑：抓取 48H 链上/盘口数据，AI 自动推演流动性池 (Liquidity Sweep) 与订单块，输出高盈亏比战术预案。
  • ⚡️ 极速自然语言执行：支持自然语言开仓（如：“做空 1000U 的 BTC”），自动测算仓位并换算 U 本位价值。
  • 🛡️ 上帝视角独立风控：彻底告别“裸奔”！智能识别多空方向，独立为现有仓位极速补挂 Algo 条件单（止盈/止损），防范插针。
  • 🏦 彭博级金库终端：一键 /balance，直接在 Telegram 渲染树状多空持仓明细与实时盈亏，账户状况一目了然。
  • 💥 一键紧急熔断：发送 /closeall，兼容单/双向持仓的三段式暴力平仓引擎瞬间接管，平仓后自动生成详尽的“战损结算清单”。
  • 🤖 自动化交易：发送‘开启自动交易’命令，启动全自动量化轮询引擎（策略无人值守执行），策略可随时优化。

📁 核心架构蓝图 (Project Structure)

我们采用了严谨的模块化分层架构，确保数据抓取、信号推演与实盘风控的绝对隔离：

binance-Quant-Zero/
├── config/                # 静态策略与全局配置 (symbols.yaml等)
├── src/                   # 量化底座引擎核心 (Data, Signals, Risk, Execution)
├── logs/                  # 运行日志存储 (自适应路径，已在 gitignore 中隔离)
├── telegram_gateway.py    # 🌟 AI 语义交互主网关 (人工指挥入口)
├── main_auto_bot.py       # 🤖 全自动量化轮询引擎 (策略无人值守执行)
├── quant_engine.py        # 🔌 独立引擎主板 (开源适配版核心接驳器)
├── requirements.txt       # Python 依赖清单
├── .env.example           # 环境变量与密钥配置模板
└── README.md              # 本机甲说明书

🚀 极速安装指南 (Quick Start)

为了更安全、稳定地运行本交易机甲，并防止与您现有的 OpenClaw 或其他机器人产生消息冲突，请严格按照以下**“零冲突架构”**进行独立部署。

🤖 方式一：OpenClaw 助手一键全自动部署 (极客推荐)

如果您已经在运行 OpenClaw，可以直接向您的 OpenClaw 助手发送以下“魔法咒语”，让它为您代劳基础安装：

完成后请告诉我，我将手动配置 .env 密钥并点火起飞。

"你好，请帮我在服务器上部署『领哥量化机甲』。请依次执行：1. git clone https://github.com/lingge66/binance-Quant-Zero.git 2. 进入目录并用 python3 -m venv quant_venv 创建虚拟环境。3. 激活环境并 pip install -r requirements.txt。4. 将 .env.example 复制为 .env。完成后告诉我，如回复你.env已完成填写，请回复我是否执行启动命令：python telegram_gateway.py

💻 方式二：标准手动部署流程

第一步：申请专属的“量化机甲 BOT” (隔离冲突)

为避免消息拦截冲突，请勿使用已有的机器人 Token。

  1. 前往 Telegram 搜索 @BotFather
  2. 发送 /newbot，花 10 秒钟创建一个全新的专属机器人（例如：lingge_quant_bot）
  3. 复制获取那个崭新的、专属的 Bot Token

第二步：克隆代码与铸造独立环境

切勿污染系统全局 Python 环境，请使用 venv 创建独立隔离区：

git clone https://github.com/lingge66/binance-Quant-Zero.git
cd binance-Quant-Zero
python3 -m venv quant_venv
source quant_venv/bin/activate  # Windows 用户: quant_venv\Scripts\activate
pip install -r requirements.txt
# 安装 PM2 进程守护 (需先安装 Node.js)
npm install -g pm2

第三步：🔐 独立配置 .env 核心密钥 (最高安全优先级)

本系统是完全解耦的，所有私密配置仅留存本地。

cp .env.example .env

打开 .env 文件，注入机甲灵魂：

  • TELEGRAM_BOT_TOKEN: 填入刚在第一步申请的专属 Token
  • BINANCE_API_KEY / SECRET: 填入您的币安 API（主网或测试网）
  • LLM_API_KEY: 填入大语言模型（OpenAI / DeepSeek 等）的 API Key 和 Base URL，用于赋能机甲的语义解析大脑

⚠️ 币安 API 权限警告：
申请 API Key 时，仅勾选“允许读取”和“允许合约交易”。
绝对禁止勾选“允许提现”！ 本系统从不需要、也永远不会要求提现权限。
重要提醒：实盘API必须绑定IP,关闭提现功能。 建议测试使用模拟API:https://demo.binance.com/en/my/settings/api-management

第四步：点火启动

# 启动 Telegram 独立网关 (人工指挥所)
python telegram_gateway.py
## 建议使用 PM2 启动网关，确保 7x24 小时在线
pm2 start telegram_gateway.py --name quant-zero-gateway

# (可选) 新开一个终端窗口，启动全自动交易引擎
python main_auto_bot.py

🛡️ 国防级安全红线 (Security Guidelines)

用户的资金安全是本系统的最高准则，内置以下四道核心防线：

  • 物理隔离：所有的私密配置必须写入 .env 文件，该文件已被 .gitignore 永久拉黑，绝不会被推送到 GitHub。
  • 拔网线级全撤全平：触发 /closeall 时，优先强制撤销所有待触发挂单释放保证金，再进行三段式暴力重试平仓。
  • 防插针挂单：风控挂单严格采用币安最新 fapi/v1/algoOrder 接口，强制绑定 CONDITIONAL 和 closePosition=true 属性，彻底杜绝止损单变反向开仓。
  • 日志脱敏：任何打印到终端或输出文件的日志，敏感 API Key 均会进行遮罩处理。

📞 技术支持与声明

主架构师：lingge66 & AI 团队 Twitter：@shangdu2005 环境要求：Ubuntu/Debian Linux/macOS/Windows, Python 3.9+

免责声明：加密货币合约交易具有极高风险。本项目代码开源仅供技术交流与比赛演示。在使用实盘（Live）模式前，请务必在 Testnet（测试网）充分验证。您对自己的所有交易行为及结果负完全责任。

🌍 English Version

🌌 Quant-Zero V1.0 Full Edition

Binance AI Trading Signal & SMC Risk Control Agent

An AI-driven trading agent based on OpenClaw, designed for Binance clients and quantitative enthusiasts. This project integrates large language model semantic parsing, SMC institutional-level market analysis, and Bloomberg Terminal-like UI panels, completely revolutionizing the interactive experience of traditional quantitative robots.

🎯 Core Features

  • 🧠 SMC Institutional Advisory Brain: Captures 48H on-chain/market data, AI automatically deduces liquidity pools (Liquidity Sweep) and order blocks, outputting high risk-reward tactical plans.
  • ⚡️ Instant Natural Language Execution: Supports natural language positions (e.g., "short 1000U worth of BTC"), automatically calculates position size and converts to USD value.
  • 🛡️ God-View Independent Risk Control: Completely告别“naked trading”！ Intelligently identifies long/short directions, independently and quickly supplements Algo conditional orders (take-profit/stop-loss) for existing positions to prevent flash crashes.
  • 🏦 Bloomberg-Level Treasury Terminal: One-click /balance command directly renders tree-style long/short position details and real-time P&L in Telegram, giving a clear overview of account status.
  • 💥 One-Click Emergency Circuit Breaker: Send /closeall command, compatible with one-way and hedge mode triple-stage violent liquidation engine instantly takes over, generating detailed "battle damage settlement report" after closing positions.
  • 🤖 Automated Trading: Send 'start auto trading' command to launch fully automated quantitative polling engine (strategy runs unattended), strategies can be optimized at any time.

📁 Project Structure

We adopt a rigorous modular layered architecture to ensure absolute isolation of data capture, signal analysis, and live risk control:

binance-Quant-Zero/
├── config/                # 静态策略与全局配置 (symbols.yaml等)
├── src/                   # 量化底座引擎核心 (Data, Signals, Risk, Execution)
├── logs/                  # 运行日志存储 (自适应路径，已在 gitignore 中隔离)
├── telegram_gateway.py    # 🌟 AI 语义交互主网关 (人工指挥入口)
├── main_auto_bot.py       # 🤖 全自动量化轮询引擎 (策略无人值守执行)
├── quant_engine.py        # 🔌 独立引擎主板 (开源适配版核心接驳器)
├── requirements.txt       # Python 依赖清单
├── .env.example           # 环境变量与密钥配置模板
└── README.md              # 本机甲说明书

🚀 Quick Start

To run this trading mech more safely and stably, and prevent message conflicts with your existing OpenClaw or other bots, please strictly follow the following "zero-conflict architecture" for independent deployment.

🤖 Method 1: OpenClaw Assistant One-Click Automated Deployment (Geek Recommended)

If you're already running OpenClaw, you can directly send the following "magic spell" to your OpenClaw assistant to handle basic installation for you:

Prompt:

"Hello, please help me deploy the 'Quant-Zero Mech' on my server. Please execute in order: 1. git clone https://github.com/lingge66/binance-Quant-Zero.git 2. Enter the directory and create a virtual environment with python3 -m venv quant_venv. 3. Activate the environment and pip install -r requirements.txt. 4. Copy .env.example to .env. After completion, let me know, and I'll manually configure the secret keys."

💻 Method 2: Standard Manual Deployment Process

Step 1: Apply for a Dedicated "Quant Mech BOT" (Isolation from Conflicts)

To avoid message interception conflicts, do not use existing bot tokens.

  1. Go to Telegram and search for @BotFather
  2. Send /newbot, spend 10 seconds creating a brand new dedicated bot (e.g., lingge_quant_bot)
  3. Copy the obtained fresh, dedicated Bot Token

Step 2: Clone Code & Create Independent Environment

Do not pollute the system-wide Python environment; use venv to create an independent isolation zone:

git clone https://github.com/lingge66/binance-Quant-Zero.git
cd binance-Quant-Zero
python3 -m venv quant_venv
source quant_venv/bin/activate  # Windows users: quant_venv\Scripts\activate
pip install -r requirements.txt
# 安装 PM2 进程守护 (需先安装 Node.js)
npm install -g pm2

Step 3: 🔐 Independent .env Core Key Configuration (Highest Security Priority)

This system is completely decoupled; all private configurations remain only locally.

cp .env.example .env

Open the .env file and inject the mech's soul:

  • TELEGRAM_BOT_TOKEN: Fill in the dedicated Token obtained in Step 1
  • BINANCE_API_KEY / SECRET: Fill in your Binance API (mainnet or testnet)
  • LLM_API_KEY: Fill in the large language model (OpenAI / DeepSeek, etc.) API Key and Base URL, empowering the mech's semantic analysis brain

⚠️ Binance API Permission Warning:
When applying for API Key, only check "Enable Reading" and "Enable Futures Trading".
Absolutely forbid checking "Enable Withdrawal"! This system never requires, and will never require withdrawal permissions.
Important reminder: Live API must bind IP, disable withdrawal function.

Step 4: Ignition Start

# Start Telegram independent gateway (manual command center)
python telegram_gateway.py
# 建议使用 PM2 启动网关，确保 7x24 小时在线
pm2 start telegram_gateway.py --name quant-zero-gateway

# (Optional) Open a new terminal window, start the fully automated trading engine
python main_auto_bot.py

🛡️ Defense-Grade Security Guidelines

User fund safety is the highest principle of this system, with the following four core defenses built-in:

  • Physical Isolation: All private configurations must be written into the .env file, which has been permanently blacklisted by .gitignore and will never be pushed to GitHub.
  • Pull-the-Plug Full Cancel & Close: When triggering /closeall, priority is given to forcibly canceling all pending conditional orders to release margin, followed by triple-stage violent retry liquidation.
  • Flash-Crash Proof Order Placement: Risk control orders strictly use Binance's latest fapi/v1/algoOrder interface, mandating CONDITIONAL and closePosition=true attributes, completely preventing stop-loss orders from becoming reverse opening positions.
  • Log Desensitization: Any logs printed to terminal or output files will have sensitive API Keys masked.

📞 Technical Support & Disclaimer

Lead Architect: lingge66& AI Team Twitter：@shangdu2005 Environment Requirements: Ubuntu/Debian Linux/macOS/Windows, Python 3.9+

Disclaimer: Cryptocurrency futures trading carries extremely high risk. This project's code is open source for technical exchange and competition demonstration only. Before using Live mode, you must fully verify on Testnet. You are solely responsible for all your trading actions and results.

Copyright (c) 2026 lingge66. All rights reserved. Licensed under the GPL-3.0 License.

About

No description, website, or topics provided.

Resources

Readme

License

GPL-3.0 license
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 99.9%
  • Shell 0.1%