← Back to Skills Marketplace
jiayinian

jy-stock-analysis

by jiayinian · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
147
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install jy-stock-analysis
Description
基于恒生聚源金融数据库,提供个股实时行情、财务分析、估值、资金流向、技术指标及机构评级的全面分析与诊断。
README (SKILL.md)

个股分析(jy-stock-analysis)

基于恒生聚源金融数据库 (MCP) 进行全面的个股分析与诊断,覆盖企业营收、行业动态、概念热点、市场行情四大维度。

重要声明:本 Skill 仅提供客观数据分析,不构成任何投资建议。股市有风险,投资需谨慎。


功能范围

模块 功能描述
实时行情 最新价、涨跌幅、成交量、市值等(强制二次核验)
公司经营 主营构成、财务指标趋势、重大事项
行业动态 行业分类、指数表现、产业链分析
概念热点 概念板块、机构评级、盈利预测、目标价
行情分析 估值分析、资金流向、技术指标

触发场景

  • "分析一下宁德时代" / "中际旭创值得投资吗"
  • "贵州茅台的估值水平" / "腾讯控股财务数据"
  • "300750 这只股票怎么样"

查询建议

查询要素:股票名称或代码(如"宁德时代"或"300750")

查询写法

  • 直接给出股票名称:分析一下比亚迪
  • 给出股票代码:看看 002594
  • 指定分析维度:分析一下茅台的估值水平

查询示例

分析一下宁德时代
帮我看看贵州茅台的财务数据
300750 这只股票怎么样
腾讯控股的估值水平如何

环境检查与配置

每次使用前必须检查 mcporter 安装和 MCP 服务配置!

步骤 1:检查 mcporter 是否安装

mcporter --version

如未安装

npm install -g mcporter
mcporter --version

步骤 2:检查 MCP 服务配置

mcporter list

预期输出(必须包含):

  • jy-financedata-tool
  • jy-financedata-api

如未配置,获取 JY_API_KEY 并配置:

2.1 获取 JY_API_KEY

向恒生聚源申请(首次配置,配置一次即可):

  • 申请邮箱[email protected]
  • 邮件标题:数据地图 KEY 申请 -XX 公司 - 申请人姓名
  • 正文:姓名、手机号、公司全称、部门、岗位、申请用途、Skill 列表

2.2 配置 MCP 服务

mcporter config add jy-financedata-tool --url "https://api.gildata.com/mcp-servers/aidata-assistant-srv-tool?token=你的 JY_API_KEY"
mcporter config add jy-financedata-api --url "https://api.gildata.com/mcp-servers/aidata-assistant-srv-api?token=你的 JY_API_KEY"

2.3 验证配置

mcporter list

步骤 3:OpenClaw 配置(如未配置)

mcporter 配置路径

  • Windows: C:\Users\用户名\config\mcporter.json

编辑 openclaw.json

{
  "skills": {
    "entries": {
      "mcporter": {
        "enabled": true,
        "env": { "MCPORTER_CONFIG": "C:\\Users\\用户名\\config\\mcporter.json" }
      }
    }
  }
}

重启 OpenClawopenclaw gateway restart


MCP 工具调用参考

工具名 功能 调用示例
FinQuery 金融问数,自然语言查询金融数据 mcporter call jy-financedata-tool.FinQuery query="查询内容"
MacroIndustryData 宏观行业经济数据 mcporter call jy-financedata-tool.MacroIndustryData query="查询内容"
FinancialResearchReport 研究报告全文摘要 mcporter call jy-financedata-tool.FinancialResearchReport query="股票代码"
FundMultipleFactorFilter 智能选基 mcporter call jy-financedata-tool.SmartFundSelection query="筛选条件"
StockMultipleFactorFilter 智能选股 mcporter call jy-financedata-tool.SmartStockSelection query="筛选条件"
AShareLiveQuote A 股实时行情 mcporter call jy-financedata-api.AShareLiveQuote query="600570"
MainOperIncData 主营构成 mcporter call jy-financedata-api.MainOperIncData query="600570"
StockValueAnalysis 股票价值分析(PE/PB/ROE 等) mcporter call jy-financedata-api.StockValueAnalysis query="600570"
StockBelongIndustry 所属行业分类 mcporter call jy-financedata-api.StockBelongIndustry query="600570"
StockBelongConcept 所属概念板块 mcporter call jy-financedata-api.StockBelongConcept query="600570"
InstitutionalRating 机构评级 mcporter call jy-financedata-api.InstitutionalRating query="600570"
ConsensusExpectationDetail 一致预期明细 mcporter call jy-financedata-api.ConsensusExpectationDetail query="600570"
PerformanceExpress 业绩快报 mcporter call jy-financedata-api.PerformanceExpress query="600570"
PerformanceForecast 业绩预告 mcporter call jy-financedata-api.PerformanceForecast query="600570"
AStockCashFlow 个股资金流向 mcporter call jy-financedata-api.AStockCashFlow query="600570"
StockRangeQuotation 股票区间行情 mcporter call jy-financedata-api.StockRangeQuotation query="600570"
CorporateResearchViewpoints 公司研究观点 mcporter call jy-financedata-api.CorporateResearchViewpoints query="600570"
NewsPublicOpinionList 新闻舆情 mcporter call jy-financedata-api.NewsPublicOpinionList query="600570"
IndustryIndexLiveQuote 行业指数实时行情 mcporter call jy-financedata-api.IndustryIndexLiveQuote query="SW801010"
ConceptIndexLiveQuote 概念指数实时行情 mcporter call jy-financedata-api.ConceptIndexLiveQuote query="CN001"
FinancialDataComparison 同行业财务数据比较 mcporter call jy-financedata-api.FinancialDataComparison query="600570"
Top10ShareHolders 十大股东 mcporter call jy-financedata-api.Top10ShareHolders query="600570"
InstitutionInvestor 机构投资者持仓 mcporter call jy-financedata-api.InstitutionInvestor query="600570"
IndexDailyQuote 指数日行情 mcporter call jy-financedata-api.IndexDailyQuote query="000001"
StockQuoteTechIndex 股票技术分析 mcporter call jy-financedata-api.StockQuoteTechIndex query="600570"
FinancialStatement 财务报表 mcporter call jy-financedata-api.FinancialStatement query="600570"
ConsensusExpectation 一致预期 mcporter call jy-financedata-api.ConsensusExpectation query="600570"
CompanyBasicInfo 公司简介 mcporter call jy-financedata-api.CompanyBasicInfo query="600570"
StockDailyQuote 股票日行情 mcporter call jy-financedata-api.StockDailyQuote query="600570"

⚠️ 强制执行流程(每次执行前必读)

执行本技能前,必须按顺序完成以下步骤,不可跳过!

步骤 0:完整阅读技能文件(强制)

  1. 阅读主技能文件SKILL.md(本文档)

  2. 阅读所有 references 文件(按顺序):

    • references/data-validation.md — 数据核验清单
    • references/report-template.md — 报告模板格式
    • references/mcp-tools.md — MCP 工具详细说明
    • references/analysis-framework.md — 分析框架详解
  3. 确认理解以下关键点

    • 实时行情必须二次核验
    • 表格数据必须无错行错列,合计正确
    • 机构评级数量合计必须等于覆盖机构总数
    • 目标价逻辑:最高价≥平均价≥中位数≥最低价
    • 报告格式必须严格按照 template 执行
    • 免责声明不可省略

未完成上述阅读和确认,不得执行后续分析步骤!


核心工作流程

步骤 1:识别股票代码

从用户输入提取股票名称或代码,无法确定时询问用户。

步骤 2:执行五大模块分析(并发调用)

模块 关键工具
实时行情 AShareLiveQuote(二次核验)
公司经营 MainOperIncData, FinQuery, StockValueAnalysis
行业动态 StockBelongIndustry, IndustryIndexLiveQuote
概念热点 StockBelongConcept, InstitutionalRating, ConsensusExpectationDetail
行情分析 AStockCashFlow, StockRangeQuotation, StockQuoteTechIndex

步骤 3:数据核验(强制)

实时行情二次核验

  • 两次调用 AShareLiveQuote 必须一致
  • 验证:涨跌幅 = (最新价 - 昨收)/昨收×100%
  • 验证:市值 = 最新价×总股本

表格核验:无错行错列,合计正确,占比和=100%

详见 references/data-validation.md

步骤 4:生成分析报告(严格按模板执行)

报告必须包含以下模块,不可省略

  1. 实时行情表 — 数据已二次核验标注
  2. 公司经营 — 主营构成 + 财务指标趋势(最近 5 期)
  3. 行业动态 — 行业分类 + 产业链分析
  4. 概念热点 — 概念板块 + 机构评级 + 盈利预测 + 目标价 + 研报观点
  5. 行情分析 — 估值分析 + 资金流向 + 技术分析
  6. 综合分析 — 核心优势 + 主要风险 + 总结
  7. 免责声明 — 必须包含,不可省略

格式要求

  • 所有表格必须无错行错列
  • 合计/总计项必须计算正确
  • 占比之和必须为 100%(容差\x3C0.1%)
  • 金额单位统一为亿元
  • 百分比变动标注 pct

详见 references/report-template.md

步骤 5:最终核验(发布前必做)

发布报告前,必须完成以下核验清单

  • 实时行情二次核验通过
  • 所有表格无错行错列
  • 合计/总计项计算正确
  • 占比之和为 100%
  • 机构数量合计正确
  • 目标价逻辑正确(最高≥平均≥中位数≥最低)
  • 上涨空间计算准确
  • 免责声明已包含
  • 数据基准日已标注

任何一项未通过,必须修正后重新核验


快速开始

⚠️ 再次提醒:执行前已完整阅读 SKILL.md 和所有 references 文件

  • 已阅读 SKILL.md
  • 已阅读 references/data-validation.md
  • 已阅读 references/report-template.md
  • 已阅读 references/mcp-tools.md
  • 已阅读 references/analysis-framework.md

工具调用命令(所有入参均为 query)

# 实时行情(二次核验)
mcporter call jy-financedata-api.AShareLiveQuote query="股票代码"
mcporter call jy-financedata-api.AShareLiveQuote query="股票代码"

# 公司经营
mcporter call jy-financedata-api.MainOperIncData query="股票代码"         # 主营构成
mcporter call jy-financedata-tool.FinQuery query="股票名称 营收 净利润 ROE"  # 财务查询
mcporter call jy-financedata-api.StockValueAnalysis query="股票代码"      # 价值分析

# 行业动态
mcporter call jy-financedata-api.StockBelongIndustry query="股票代码"     # 行业分类
mcporter call jy-financedata-api.IndustryIndexLiveQuote query="行业代码"  # 行业指数

# 概念热点
mcporter call jy-financedata-api.StockBelongConcept query="股票代码"          # 概念板块
mcporter call jy-financedata-api.InstitutionalRating query="股票代码"         # 机构评级
mcporter call jy-financedata-api.ConsensusExpectationDetail query="股票代码"  # 一致预期

# 行情分析
mcporter call jy-financedata-api.AStockCashFlow query="股票代码"         # 资金流向
mcporter call jy-financedata-api.StockRangeQuotation query="股票代码"    # 区间行情
mcporter call jy-financedata-api.StockQuoteTechIndex query="股票代码"    # 技术分析

# 研报观点
mcporter call jy-financedata-tool.FinancialResearchReport query="股票代码"    # 研报摘要
mcporter call jy-financedata-api.CorporateResearchViewpoints query="股票代码" # 研究观点
mcporter call jy-financedata-api.NewsPublicOpinionList query="股票代码"       # 新闻舆情

数据获取策略

  1. 并发调用:五大模块尽量并发执行
  2. 实时行情优先:先核验基础数据
  3. 按需调用:根据用户关注点调整
  4. 失败重试:单个失败重试 1-2 次

资源清单

skills/jy-stock-analysis/
├── SKILL.md
├── references/
│   ├── execution-checklist.md  # 执行清单(强制执行)
│   ├── data-validation.md      # 数据核验清单
│   ├── report-template.md      # 报告模板
│   ├── mcp-tools.md            # 工具说明
│   └── analysis-framework.md   # 分析框架
└── .learnings/                 # 运行日志(可选,用于记录执行历史)

重要:每次执行前必须阅读 references/execution-checklist.md,按步骤逐项完成!


输出格式要求

  • 财务表格:最近 5 期数据,含同比变化
  • 金额单位:亿元;百分比变动:pct
  • 机构评级:机构总数、各评级数量/占比、最新日期
  • 目标价:最高/最低/平均/中位数、上涨空间
  • 研报观点:至少 3-5 篇,日期倒序

注意事项

  1. 数据时效:行情实时/T+1,财务为最新报告期
  2. 行业分类:使用申万行业分类
  3. 估值选择:高 PE 行业用 PE,低 PB 行业用 PB
  4. 免责声明:必备,不可省略
  5. 预测标注:明确预测截止日期和最新评级日期
  6. 强制阅读:每次执行前必须完整阅读 SKILL.md 和所有 references 文件
  7. 数据核验:必须完成步骤 5 的最终核验清单才能发布报告

执行检查清单(每次执行必做)

执行前确认

  • 已完整阅读 SKILL.md
  • 已完整阅读 references/data-validation.md
  • 已完整阅读 references/report-template.md
  • 已完整阅读 references/mcp-tools.md
  • 已完整阅读 references/analysis-framework.md
  • 理解数据核验要求
  • 理解报告格式要求

执行中确认

  • 实时行情已二次核验
  • 表格数据已核验(无错行错列)
  • 合计/总计项已验算
  • 占比之和为 100%

发布前确认

  • 完成步骤 5 最终核验清单
  • 免责声明已包含
  • 数据基准日已标注

限制

  1. 市场限制:仅支持 A 股,港股/美股需其他工具
  2. 行情:非交易时段为最近收盘价
  3. 财务:最新已披露报告期
  4. 投资建议:不提供买卖建议
  5. API 频率:注意调用频率限制

故障排查

问题 原因 解决
mcporter 不存在 未安装 npm install -g mcporter
服务未找到 未配置 mcporter list 检查后配置
KEY 无效 过期/错误 重新申请
数据为空 代码错误/停牌 确认代码正确
调用超时 网络/限流 稍后重试
工具不存在 服务配置错误 检查 MCP 服务配置是否正确

检查清单

  • mcporter 已安装
  • 两个 MCP 服务已配置
  • JY_API_KEY 有效
  • OpenClaw 已重启
  • 股票代码正确(6 位数字)
  • 工具调用路径正确

版本历史

V1.0.0 (2026-04-01)

  • 初始版本发布
  • 基于恒生聚源金融数据库 (MCP) 提供全面的个股分析功能
  • 覆盖实时行情、公司经营、行业动态、概念热点、行情分析五大模块
  • 强制数据核验机制:实时行情二次核验、表格结构核验、逻辑一致性检查
  • 标准化报告模板:7 大模块 + 免责声明
  • 完整执行清单:执行前/执行中/发布前三阶段确认
Usage Guidance
Key points to check before installing/using this skill: - Metadata mismatch: SKILL.md expects node/npm and the mcporter tool and instructs installing mcporter via npm, but the registry entry lists no required binaries or env vars. Do not trust the absence of declared secrets — SKILL.md actually requires an MCP API key (JY_API_KEY). - Secret handling: The instructions tell you to embed your JY_API_KEY in mcporter config URLs (token as a query parameter) and to store MCPORTER_CONFIG in openclaw.json. That will persist your API key on disk and could expose it to anyone with access to those files. Prefer using a secure secret store or review mcporter's config encryption/permissions before storing tokens. - Verify mcporter: Research the 'mcporter' npm package and its publisher. Only install packages from trusted sources and inspect their code (or vendor docs) because installing a global npm package runs code on your machine. - Config edits and restart: The skill asks you to edit openclaw.json and restart the OpenClaw gateway. If you are not an admin, ask an administrator to perform and audit these changes. Confirm that enabling mcporter in your agent does not inadvertently grant external code broader access than intended. - Least privilege: Request an MCP key scoped with the minimal permissions necessary and rotate/revoke it if you stop using the skill. Avoid reuse of more-privileged keys. - If you proceed: run mcporter and any installs in a restricted/sandbox environment first, verify network endpoints (api.gildata.com), and confirm that the mcporter package and any generated config files do not exfiltrate data to unknown endpoints. If you want, I can list specific checks you (or an admin) should perform on the mcporter package and on openclaw.json before enabling the skill.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The skill's purpose — querying 恒生聚源 (MCP) for stock analysis — matches the runtime instructions that call mcporter and many MCP endpoints. However, SKILL.md metadata declares required binaries (node, npm, mcporter) and an npm install for mcporter, while the registry metadata at the top lists no required binaries or installs. This mismatch between the skill's internal SKILL.md and the published registry metadata is an incoherence to verify.
Instruction Scope
Runtime instructions require running mcporter commands, obtaining a JY_API_KEY, configuring mcporter with a URL that embeds the token, and editing openclaw.json to enable the mcporter entry and set MCPORTER_CONFIG — including restarting the OpenClaw gateway. These are functional for the stated purpose, but they ask the user to store an API key in a config URL and to directly modify agent configuration and restart services, which expands the skill's operational footprint and can expose secrets if done carelessly. Also SKILL.md enforces many mandatory manual verification steps which grant the skill broad discretion in what it reads/uses as 'context'.
Install Mechanism
The published package contains no install spec and no code files (instruction-only), which is low-risk in general. But SKILL.md includes an install hint to 'npm install -g mcporter' (and lists node/npm/mcporter as required bins). That implied install-from-npm is reasonable if mcporter is an official tool, but the registry metadata did not declare this. Confirm the legitimacy of the mcporter npm package and review its source before installing.
Credentials
The instructions require obtaining and using a JY_API_KEY (MCP token) and embedding it into mcporter config URLs, but the skill's registry metadata declares no required environment variables or primary credential. This is a clear omission: a secret is needed at runtime but not declared. Storing API tokens in plaintext config files or URLs is risky; the skill also suggests setting MCPORTER_CONFIG via openclaw.json which may persist the token on disk.
Persistence & Privilege
The skill is not marked always:true and is user-invocable. However, it instructs modifying OpenClaw agent configuration (openclaw.json) to enable mcporter and set MCPORTER_CONFIG and to restart the gateway. Editing agent config is within the plausible setup for enabling an external tool, but it is a higher-privilege action than a purely ephemeral skill and should be reviewed by an administrator before applying.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install jy-stock-analysis
  3. After installation, invoke the skill by name or use /jy-stock-analysis
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
jy-stock-analysis 1.0.1 - Updated tool names in the MCP tools reference table for "智能选基" and "智能选股". - No changes to functionality or workflow; documentation update only.
v1.0.0
jy-stock-analysis 1.0.0 - Initial release providing comprehensive stock analysis powered by 恒生聚源金融数据库 (MCP). - Supports real-time quotes (with mandatory double-check), financials (DuPont framework), valuation, capital flow, technical indicators, institutional ratings, and earnings forecasts. - Covers company operations, industry trends, concept hot topics, and market analysis. - Detailed usage instructions, environment setup, and mandatory validation workflows included. - Objective data analysis only; no investment advice provided.
Metadata
Slug jy-stock-analysis
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is jy-stock-analysis?

基于恒生聚源金融数据库,提供个股实时行情、财务分析、估值、资金流向、技术指标及机构评级的全面分析与诊断。 It is an AI Agent Skill for Claude Code / OpenClaw, with 147 downloads so far.

How do I install jy-stock-analysis?

Run "/install jy-stock-analysis" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is jy-stock-analysis free?

Yes, jy-stock-analysis is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does jy-stock-analysis support?

jy-stock-analysis is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created jy-stock-analysis?

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

💬 Comments