← 返回 Skills 市场
youngcan-wang

Wyckoff Agent Skill

作者 YoungCan-Wang · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
89
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install wyckoff-agent-skill
功能描述
Wyckoff A-share analysis agent with full CLI integration. Detects local CLI installation, guides users through setup (install → register → configure data sou...
安全使用建议
This skill's functionality (Wyckoff analysis via a CLI) is coherent, but exercise caution before installing or running commands it suggests: 1) Do not run the curl | bash one-liner unless you inspect the install.sh content in the GitHub repo and confirm the repo/author are trustworthy. 2) Prefer installing the package in an isolated environment (virtualenv, container) and inspect its code before running. 3) Be aware the CLI persists credentials to ~/.wyckoff/wyckoff.json and may auto-relogin; if you supply API keys or account passwords, consider using least-privilege keys or dedicated/test accounts. 4) Verify the pip package (youngcan-wyckoff-analysis) exists on PyPI and review its project homepage, code, and recent activity; absence of a homepage/source in the registry is a red flag. 5) If you need higher assurance, ask the publisher for the canonical repository, a vetted install artifact (PyPI release), and a copy of the install script for manual review — providing those would increase confidence and could move the assessment toward 'benign'.
功能分析
Type: OpenClaw Skill Name: wyckoff-agent-skill Version: 1.0.1 The skill bundle promotes high-risk environment setup procedures, including a 'curl | bash' one-liner and 'pip install' for a specific third-party package (youngcan-wyckoff-analysis) in SKILL.md and rules/cli-setup-guide.md. It explicitly guides users to provide sensitive credentials and API keys for financial data services (Tushare, TickFlow), which are stored in a local config file (~/.wyckoff/wyckoff.json). While the logic is consistent with its stated purpose as a trading assistant, the inclusion of a hardcoded affiliate referral link (tickflow.org/auth/register?ref=5N4NKTCPL4) and the reliance on external script execution from GitHub (raw.githubusercontent.com/YoungCan-Wang/Wyckoff-Analysis/main/install.sh) present significant supply chain and RCE risks without clear evidence of intentional malice.
能力标签
cryptocan-make-purchasesrequires-sensitive-credentials
能力评估
Purpose & Capability
Name/description (Wyckoff A-share CLI-based analysis) matches the instructions: install/initialize a wyckoff CLI, configure data sources (Tushare/TickFlow), set an LLM model provider, and run analysis. Required capabilities (time, web fetch, CSV/image parsing, plotting) are coherent with the stated purpose.
Instruction Scope
SKILL.md explicitly instructs running CLI commands, performing online fetches, reading CSV/images, and persisting credentials. Those actions are relevant, but the instructions also direct the user/agent to run remote install commands and to persist login tokens automatically—operations that can write files and perform network auth without clarifying storage/encryption or verifying the install script contents.
Install Mechanism
The registry contains no formal install spec, but SKILL.md recommends 'pip install youngcan-wyckoff-analysis' and offers a one-line 'curl -fsSL https://raw.githubusercontent.com/YoungCan-Wang/Wyckoff-Analysis/main/install.sh | bash'. Piping a remote raw script into bash is high risk unless the script and repository are vetted. The pip package and GitHub repo appear unverified/unknown in metadata (homepage/source unknown), increasing install risk.
Credentials
The skill expects users to configure service API keys (Tushare token, TickFlow API key) and model provider API keys (gemini/openai/claude). Those credentials are proportionate to the functionality. However, the skill does not declare env var requirements up front and documents that credentials are persisted to ~/.wyckoff/wyckoff.json and that automatic re-login can occur — this persistent local storage of secrets should be considered before installing.
Persistence & Privilege
always:false (no forced inclusion) and model invocation is allowed (normal). The skill's flow instructs storing tokens on disk and automatic re-login, which grants ongoing network/auth activity and persisted secrets on the host. Combined with the remote-install recommendation and unknown origin, this persistent behavior increases the blast radius if the installed CLI or its install script are malicious or compromised.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install wyckoff-agent-skill
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /wyckoff-agent-skill 触发
  4. 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
Wyckoff-agent-skill 1.0.1 - Added comprehensive SKILL.md documentation detailing setup, onboarding, and operational workflows. - Clarified two operating modes: guided setup (environment checks) and full Wyckoff-style analysis pipeline. - Documented all input protocols, workflow steps (0–9), output contract, and hard technical constraints. - Expanded support for both CLI portfolio management commands and multi-modal inputs (symbols, holdings, CSV, images). - Included explicit data source fallback rules and capability degradation handling.
元数据
Slug wyckoff-agent-skill
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Wyckoff Agent Skill 是什么?

Wyckoff A-share analysis agent with full CLI integration. Detects local CLI installation, guides users through setup (install → register → configure data sou... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 89 次。

如何安装 Wyckoff Agent Skill?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install wyckoff-agent-skill」即可一键安装,无需额外配置。

Wyckoff Agent Skill 是免费的吗?

是的,Wyckoff Agent Skill 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Wyckoff Agent Skill 支持哪些平台?

Wyckoff Agent Skill 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 Wyckoff Agent Skill?

由 YoungCan-Wang(@youngcan-wang)开发并维护,当前版本 v1.0.1。

💬 留言讨论