← 返回 Skills 市场
845
总下载
1
收藏
1
当前安装
2
版本数
在 OpenClaw 中安装
/install vibetrading
功能描述
Build, backtest, and deploy cryptocurrency trading strategies using the vibetrading Python framework. Use when: (1) generating trading strategies from natura...
安全使用建议
This skill's instructions look like a legitimate crypto trading framework, but there are two red flags you should resolve before installing or handing over secrets: (1) the registry metadata does not declare the environment variables the SKILL.md clearly needs (exchange API keys / private keys, and ANTHROPIC_API_KEY or OPENAI_API_KEY). That mismatch could be an oversight or a sign the publisher omitted sensitive requirements. (2) There is no homepage or source URL listed — you have no way to review the package source or verify the maintainer. Recommended actions: verify the vibetrading PyPI package and its source repository (look for a homepage, GitHub repo, and recent commits); prefer exchange API keys with minimal scopes or test with read-only/sandbox credentials first; never paste a long-term private key into a project you haven't audited (use ephemeral keys or hardware wallets where supported); require the publisher to add explicit required env metadata and a source link before using for live trading; and run all code in isolated sandboxes/backtests until you confirm behavior.
功能分析
Type: OpenClaw Skill
Name: vibetrading
Version: 1.0.1
The vibetrading skill bundle provides a framework for crypto trading that involves high-risk behaviors, specifically the handling of sensitive exchange private keys and API secrets (SKILL.md, references/api-details.md). It also includes a workflow for generating and executing Python code via LLMs (vibetrading.strategy.generate), which introduces a significant risk of code injection or unintended execution. Furthermore, the metadata and documentation contain future-dated timestamps and model versions (e.g., publishedAt in 2026 and 'claude-sonnet-4-20250514'), which are anomalous indicators.
能力评估
Purpose & Capability
The skill's name and prose describe building, backtesting, and deploying crypto strategies and the SKILL.md provides exactly those instructions (backtest API, live deploy, exchange credentials, AI generation). However the registry metadata shows no required env vars or homepage/source, which is inconsistent with the clear runtime need for exchange API keys and LLM API keys.
Instruction Scope
Instructions are narrowly scoped to authoring strategies, backtesting, and live deployment. They explicitly tell the agent to read strategy files, load a .env.local for credentials, call vibetrading.live.start(...) with api_key/api_secret, and optionally use ANTHROPIC_API_KEY or OPENAI_API_KEY for AI generation. The only scope issue is that these runtime behaviors require sensitive secrets and file reads but the skill metadata does not declare them.
Install Mechanism
This is an instruction-only skill with no install spec and no bundled code; SKILL.md recommends pip install vibetrading (normal for a Python library). No remote download/extract or unknown install host is specified in the skill bundle itself.
Credentials
The runtime clearly needs exchange credentials (wallet/private key or api key/secret) and LLM API keys for AI generation, yet the skill metadata lists no required env vars and no primary credential. The absence of declared env requirements plus the suggestion to store private keys in .env.local is a mismatch and increases risk — you should only provide minimal credentials, prefer scoped API keys, and verify where keys are used.
Persistence & Privilege
always:false and no install steps mean the skill does not request permanent elevated presence. It does instruct reading local files (strategy code, .env.local) and using those values at runtime, which is expected for live trading. No instructions attempt to change 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 vibetrading - 安装完成后，直接呼叫该 Skill 的名称或使用
/vibetrading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
v1.0.1
Version 1.0.1
- Major project cleanup: removed legacy example scripts, templates, and outdated validation/backtest files.
- Updated SKILL.md to document the new vibetrading Python framework with a focus on function-based strategy design, backtesting, and live trading.
- Added a concise API reference file (references/api-details.md) for easier strategy and function lookup.
- Skill is now agent-first and supports multi-exchange deployment (Hyperliquid, Paradex, Lighter, Aster).
- Skill is strictly limited to crypto trading strategy generation, backtesting, and deployment within the vibetrading framework.
v1.0.0
Initial release of vibetrading for clawdbot(https://vibetrading.dev)
- Generate executable Hyperliquid trading strategy code from natural language prompts.
- Produces complete Python code, including error handling, logging, config management, and Hyperliquid API integration.
- Includes an automatic code validation and fixing system for Python 3.5+ compatibility.
- Supports a range of trading strategies: technical indicator, grid, mean reversion, trend following, signal-based, and more.
- Offers command-line tools for code generation and validation, with flexible configuration options.
- Ensures security (no hardcoded credentials) and adherence to code quality standards.
元数据
常见问题
Vibetrading 是什么？
Build, backtest, and deploy cryptocurrency trading strategies using the vibetrading Python framework. Use when: (1) generating trading strategies from natura... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 845 次。
如何安装 Vibetrading？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vibetrading」即可一键安装，无需额外配置。
Vibetrading 是免费的吗？
是的，Vibetrading 完全免费（开源免费），可自由下载、安装和使用。
Vibetrading 支持哪些平台？
Vibetrading 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 Vibetrading？
由 crabbytt（@crabbytt）开发并维护，当前版本 v1.0.1。
推荐 Skills