← 返回 Skills 市场
Alpha Hound
作者
zhengxinjipai
· GitHub ↗
· v0.1.0
302
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install alpha-hound-cn
功能描述
基于复杂条件筛选股票并输出候选列表。
使用说明 (SKILL.md)
Alpha Hound
目标
根据复杂筛选条件深度查询并筛选股票，输出候选与理由。
输入
- 条件集合（财务、估值、技术、行业、事件）
- 市场范围（美股/港股/全球）
- 时间窗口与更新频率
方法
- 解析条件：将自然语言条件拆为可查询字段
- FinnHub API 筛选：基本面与财务指标
- SERPER_API 搜索：新闻/事件/行业热点
- JINA_API 抽取：提取网页核心证据
- 交叉验证：确认条件满足度与风险
输出格式
- 筛选条件摘要
- 候选列表（含满足条件的证据）
- 风险与不确定性
- 建议的下一步验证
- 要点表格（Markdown Table）
安全使用建议
Do not install or enable this skill until the author clarifies how external API credentials are provided and why the Read tool is permitted. Ask for (1) explicit required env var names (e.g., FINNHUB_API_KEY, SERPER_API_KEY, JINA_API_KEY) and a justification for any local file access, (2) whether the skill will store or transmit query/selection data to any endpoints beyond the named APIs, and (3) why the skill is not user-invocable but can be run autonomously. If credentials are to be supplied by the platform, verify least-privilege scoping and that secrets are not exposed in logs or sent to untrusted hosts.
功能分析
Type: OpenClaw Skill
Name: alpha-hound-cn
Version: 0.1.0
The Alpha Hound skill bundle is a financial analysis tool designed for stock screening based on user-defined criteria. The SKILL.md and skill.json files outline a legitimate workflow using standard APIs (FinnHub, Serper, Jina) and tools (WebSearch, WebFetch) to gather and process market data, with no evidence of malicious intent, data exfiltration, or prompt injection attacks.
能力评估
Purpose & Capability
SKILL.md explicitly names FinnHub, SERPER_API, and JINA_API as data sources for screening stocks, but the skill metadata declares no required environment variables or primary credential. A skill that depends on external paid APIs would normally require API keys; the absence of declared credentials is incoherent.
Instruction Scope
Runtime instructions allow use of Read, WebSearch, and WebFetch. WebFetch/WebSearch match the stated web-sourcing purpose, but the Read tool could permit arbitrary local file access while the SKILL.md does not justify reading local files. The instructions also do not say how credentials will be supplied or whether retrieved data (search queries, extracted evidence) is sent to external endpoints beyond the named APIs.
Install Mechanism
This is an instruction-only skill with no install spec or code files, so nothing will be downloaded or written during install. That limits direct supply-chain risk.
Credentials
The skill will call three external services but declares zero required env vars. Expected variables like FINNHUB_API_KEY, SERPER_API_KEY, or JINA_API_KEY are missing. Asking for no credentials while planning to use third-party APIs is disproportionate and ambiguous.
Persistence & Privilege
always is false (good). user-invocable is false while disable-model-invocation is false, meaning the agent could call the skill autonomously but users can't directly invoke it—this is unusual and worth confirming but not sufficient alone to downgrade further.
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 alpha-hound-cn - 安装完成后，直接呼叫该 Skill 的名称或使用
/alpha-hound-cn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
v0.1.0
Initial release of alpha-hound skill:
- Enables advanced stock screening based on complex, customizable conditions.
- Supports filtering by financials, valuation, technicals, sector, and events.
- Integrates multiple APIs for data extraction, news/event search, and cross-verification.
- Outputs candidate stock lists with evidence, risk assessment, and actionable next steps.
- Delivers concise results with tabular summaries in Markdown format.
元数据
常见问题
Alpha Hound 是什么？
基于复杂条件筛选股票并输出候选列表。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 302 次。
如何安装 Alpha Hound？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install alpha-hound-cn」即可一键安装，无需额外配置。
Alpha Hound 是免费的吗？
是的，Alpha Hound 完全免费（开源免费），可自由下载、安装和使用。
Alpha Hound 支持哪些平台？
Alpha Hound 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 Alpha Hound？
由 zhengxinjipai（@zhengxinjipai）开发并维护，当前版本 v0.1.0。
推荐 Skills