← Back to Skills Marketplace
chenghaifeng08-creator

Autonomous Trading System

by chenghaifeng08-creator · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
140
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install autonomous-trading-system-automaton
Description
全自动智能交易系统 - 无人值守、自我进化、稳定盈利。包含风险控制、市场状态识别、动态止损、仓位管理等核心功能。
README (SKILL.md)

🤖 Autonomous Trading System

全自动智能交易系统 - 真正无人值守、自我进化、稳定盈利


💰 付费服务

交易策略咨询 & 定制:

服务 价格 说明
策略回测验证 ¥1000/策略 历史数据回测 + 风险评估
定制交易系统 ¥5000 起 根据你的需求定制
1 对 1 交易指导 ¥2000/小时 仓位管理 + 心态辅导
月度顾问 ¥8000/月 每周策略调整 + 每日监控

⚠️ 重要: 交易有风险,不承诺收益。我们提供工具和策略,决策由你负责。

联系: 微信/Telegram 私信,备注"交易咨询"


核心功能

1. 风险控制

  • 单笔风险限制 (\x3C2% 总资金)
  • 日亏损熔断 (-$10 停止交易)
  • 最大持仓数限制
  • 黑名单机制

2. 市场状态识别

  • 趋势市检测
  • 震荡市检测
  • 波动率评估
  • 动态调整策略

3. 智能止损

  • ATR 动态止损
  • 移动止盈
  • 时间止损
  • 分批退出

4. 仓位管理

  • Kelly 公式优化
  • 风险平价分配
  • 相关性对冲
  • 杠杆控制

5. 自我进化

  • 交易表现分析
  • 策略自动优化
  • 参数自适应
  • 学习业界最佳实践

使用示例

# 启动交易系统
node auto-trading-bot.js start

# 查看持仓
node auto-trading-bot.js positions

# 手动触发优化
node auto-trading-bot.js optimize

# 查看性能报告
node auto-trading-bot.js report

配置

{
  "maxPositions": 5,
  "riskPerTrade": 0.02,
  "dailyStopLoss": -10,
  "checkInterval": 300000
}

注意事项

⚠️ 高风险警告: 加密货币交易存在重大风险,可能导致本金全部损失。

⚠️ 历史教训: v13.0 系统 3 天亏损 -24.1%,主要原因是交易过频和策略缺陷。

⚠️ 建议: 先用小额资金测试,确认策略有效后再增加投入。

文件结构

autonomous-trading-system/
├── SKILL.md              # 技能说明
├── ARCHITECTURE.md       # 系统架构文档
├── auto-trading-bot.js   # 主交易机器人
├── risk-manager.js       # 风险管理
├── market-analyzer.js    # 市场分析
└── optimizer.js          # 策略优化

版本历史

  • v14.0 (2026-03-19): 完全重构,吸取 v13.0 失败教训
  • v13.0 (2026-03-14): 初始版本,3 天后因策略缺陷停止

作者: chenghaifeng08-creator
许可证: MIT
最后更新: 2026-03-19

Usage Guidance
Do not install or run this skill as-is. Key issues: (1) The SKILL.md expects Node scripts that are not included in the package, so the skill is non-functional or relies on fetching code from elsewhere — ask the publisher for the actual source files and a clear install procedure. (2) The metadata inconsistently lists required environment variables (BINANCE API keys) while other parts show none; if you supply exchange API keys only provide restricted keys (disable withdrawals) and use least privilege. (3) Verify the author's identity and hosting/source (GitHub, release tarball) and review the actual code before giving any credentials or funds. Recommended next steps: request a full source bundle or a link to a trusted repository, confirm a documented install method, insist on explicit guidance for API key scopes, run the code in an isolated environment and test with minimal funds or paper trading. If the author asks you to send credentials via private messaging (WeChat/Telegram) or to run code fetched from ad-hoc URLs, treat that as high risk.
Capability Analysis
Type: OpenClaw Skill Name: autonomous-trading-system-automaton Version: 1.0.1 The skill bundle is classified as suspicious primarily due to the inclusion of a 'Paid Services' section in SKILL.md, which solicits off-platform payments via WeChat/Telegram for trading consulting and 'stable profit' strategies—common indicators of financial scams. While ARCHITECTURE.md provides detailed pseudocode for a trading system, the actual implementation files (e.g., auto-trading-bot.js, risk-manager.js) are missing from the bundle, making it impossible to verify the safety of the executable code. The combination of high-risk financial claims and the solicitation of external payments suggests a potential social engineering or scam risk rather than a direct technical exploit.
Capability Assessment
Purpose & Capability
The SKILL.md describes a Node-based trading bot and shows commands like `node auto-trading-bot.js start` and references exchange usage, which logically require a node binary and exchange API credentials. However the top-level Requirements block lists no binaries or env vars, while _meta.json does list 'node' and 'BINANCE_API_KEY'/'BINANCE_API_SECRET'. This mismatch between declared requirements and actual instructions is inconsistent.
Instruction Scope
Runtime instructions direct the agent/user to run local Node scripts (start, positions, optimize, report) and include a config example, but none of the referenced JS files are present in the package. The SKILL.md does not instruct reading unrelated system files or exfiltrating data, but it does ask users to contact authors via private messaging for paid services (social-engineering risk). The lack of included code makes the instructions non-executable and potentially a prompt to fetch code from elsewhere.
Install Mechanism
There is no install spec (instruction-only), which is low risk by itself. However the skill advertises and documents JS files (auto-trading-bot.js, risk-manager.js, etc.) that are not bundled — without an install mechanism or packaged code, it's unclear how those files would appear, which is an operational/integrity concern (could indicate missing files or expectation of external downloads).
Credentials
_meta.json declares BINANCE_API_KEY and BINANCE_API_SECRET (reasonable for a trading bot), but the package's visible requirements block earlier shows no required env vars. The skill asks for sensitive exchange credentials implicitly but does not consistently declare them. Requesting exchange API keys is proportional for a trading bot, but the inconsistency and absence of guidance about least-privilege (e.g., 'withdrawals disabled' API key) is concerning.
Persistence & Privilege
The skill does not request always:true and does not declare system-wide config modifications. There are no special persistence or privilege requests in the package metadata.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install autonomous-trading-system-automaton
  3. After installation, invoke the skill by name or use /autonomous-trading-system-automaton
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added detailed documentation (SKILL.md) describing core features, usage, configuration, and service offerings. - Outlined risk controls, market state recognition, dynamic stop-loss, position management, and self-optimization capabilities. - Included example commands, sample configuration, and file structure overview. - Provided risk warnings, historical lessons, and advice for safe usage. - Summary of version history and author information added.
Metadata
Slug autonomous-trading-system-automaton
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Autonomous Trading System?

全自动智能交易系统 - 无人值守、自我进化、稳定盈利。包含风险控制、市场状态识别、动态止损、仓位管理等核心功能。 It is an AI Agent Skill for Claude Code / OpenClaw, with 140 downloads so far.

How do I install Autonomous Trading System?

Run "/install autonomous-trading-system-automaton" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Autonomous Trading System free?

Yes, Autonomous Trading System is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Autonomous Trading System support?

Autonomous Trading System is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Autonomous Trading System?

It is built and maintained by chenghaifeng08-creator (@chenghaifeng08-creator); the current version is v1.0.1.

💬 Comments