← 返回 Skills 市场
Quant Analyst
作者
Justin Liu
· GitHub ↗
· v0.1.1
580
总下载
0
收藏
9
当前安装
2
版本数
在 OpenClaw 中安装
/install quant-analyst
功能描述
Professional quantitative trading system for cryptocurrency - backtesting, paper trading, live trading, and strategy optimization
安全使用建议
Proceed cautiously. The quant-analyst instructions themselves are consistent with a trading tool, but:
- There is an unrelated SKILL-EN.md (video-generator) bundled in the same package — ask the publisher why unrelated skill docs are included.
- The skill's metadata declares no required env vars but the instructions expect BINANCE_API_KEY/SECRET (and the English file expects OPENAI_API_KEY). Treat this as a mismatch and assume the skill will ask for credentials.
- Before running any git clone / pip install or executing scripts from the repos, manually inspect the GitHub repositories (openclaw-quant and openclaw-video) and review scripts such as generate-for-openclaw.sh and any startup hooks. Check recent commits, contributors, and issues.
- Never supply exchange API keys with withdrawal permissions. Use read-only or testnet API keys and start with paper trading only.
- Run installations and live tests in an isolated environment (VM or container) until you verify the code.
- Prefer skills with a trustworthy homepage or known maintainer. If you can't verify the upstream repositories or get a clear explanation for the bundled video doc, consider not installing.
功能分析
Type: OpenClaw Skill
Name: quant-analyst
Version: 0.1.1
The bundle exhibits a critical discrepancy: while the metadata and primary documentation (SKILL.md, SKILL-ZH.md) describe a 'quant-analyst' trading tool, SKILL-EN.md contains instructions for a completely unrelated 'video-generator' skill. This inclusion of 'shadow' instructions for a different domain—including commands to clone external repositories (github.com/ZhenRobotics/openclaw-video) and execute shell scripts—constitutes a prompt-injection risk where the agent may perform actions beyond the user's intended scope. While no explicit data exfiltration logic is present, the deceptive packaging and unauthorized command execution paths are highly irregular.
能力评估
Purpose & Capability
The primary SKILL.md and SKILL-ZH.md describe a quantitative trading tool and instruct cloning https://github.com/ZhenRobotics/openclaw-quant and installing Python deps—this is coherent with the stated purpose. However, SKILL-EN.md is for a completely different 'video-generator' skill (different repo, different env var: OPENAI_API_KEY) bundled into the same package. That file is unrelated to quant-analyst and suggests a packaging/misconfiguration or intentional bundling of unrelated capabilities.
Instruction Scope
The runtime instructions tell the agent/user to git clone external repositories and run project scripts (pip install -r requirements.txt, run python -m openclaw_quant, and in the English file run ~/openclaw-video/generate-for-openclaw.sh). Those commands will fetch and execute third-party code on the user's machine. The SKILL.md examples reference reading env vars (BINANCE_API_KEY/SECRET and use of os.getenv) even though the declared skill metadata lists no required env vars. The presence of unrelated instructions (video-generator) increases risk that the agent may run commands outside the intended domain.
Install Mechanism
No formal install spec is declared (instruction-only), which is lower baseline risk for the platform, but the instructions require cloning GitHub repos and running pip/npm installs. The repos referenced are hosted on GitHub (reasonable), but cloning and running install scripts pulls arbitrary code from an external source into the user's environment and may execute scripts that are not pre-vetted. The package contains two different repo targets (openclaw-quant and openclaw-video).
Credentials
Declared metadata lists no required env vars, but the instructions and examples explicitly advise setting BINANCE_API_KEY and BINANCE_API_SECRET (for live trading) and SKILL-EN.md requests OPENAI_API_KEY. This mismatch is a red flag: the skill will operate with credentials but does not declare them, so a user or automated installer may not be warned. Live trading also implies use of exchange API keys—users must ensure keys are limited (no withdrawal) and prefer testnet keys for initial use.
Persistence & Privilege
The skill does not request always:true, does not declare special config paths, and is install-free at platform level. It can be invoked autonomously (platform default) and contains AUTO-TRIGGER keywords, which is expected for agent-facing skills. There is no evidence the skill attempts to modify other skills or system-wide settings.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install quant-analyst - 安装完成后，直接呼叫该 Skill 的名称或使用
/quant-analyst触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.1.1
No changes detected for version 0.1.1.
- No file or documentation changes in this release.
- Functionality and features remain identical to version 0.1.0.
v0.1.0
quant-analyst 0.1.0 – Initial Release
- Launches a professional quantitative trading system for cryptocurrency.
- Core features: backtesting, paper trading, live trading, and strategy optimization.
- Supports 50+ technical indicators and multiple exchanges (via ccxt).
- Includes built-in strategies (e.g., MA Cross, RSI, MACD, Bollinger).
- Offers performance analytics (Sharpe ratio, max drawdown, win rate, etc.) and risk management tools.
- Easy installation, CLI, and agent integration for natural language queries.
元数据
常见问题
Quant Analyst 是什么？
Professional quantitative trading system for cryptocurrency - backtesting, paper trading, live trading, and strategy optimization. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 580 次。
如何安装 Quant Analyst？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install quant-analyst」即可一键安装，无需额外配置。
Quant Analyst 是免费的吗？
是的，Quant Analyst 完全免费（开源免费），可自由下载、安装和使用。
Quant Analyst 支持哪些平台？
Quant Analyst 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Quant Analyst？
由 Justin Liu（@zhenstaff）开发并维护，当前版本 v0.1.1。
推荐 Skills