rootSunc/CNEquity

❌ 实测失败serverPythonApache-2.0★ 997

开源中国股市（A股）本地数据湖：沪深股票行情、A股股票数据与金融数据，面向量化研究、量化交易与回测，含财务数据、期货数据、资金流向、公告、指数行业与宏观。86个日更数据集，复权与历史成分内置，行级溯源、PIT，MCP原生。免注册、免 Token，自托管。Open-source China stock market data lake for quant research, backtesting and AI agents.

实测结果

失败

沙箱内启动/握手失败，下方给出失败阶段与原始报错。失败阶段：initialize，错误类型：runtime_error。这类信息对判断“是不是我这边的环境问题”很有用，原始报错见下。

error_type: runtime_error

Inspect historical decision-data evidence and unresolved
                 gaps.

Use the lake:
  query  Run DuckDB SQL, or fetch a single dataset on demand.
  serve  Start the lake dashboard.
  mcp    Open this lake to AI agents over MCP (stdio or HTTP).

Governance and inspection:
  snapshot      Create, verify and safely restore portable lake snapshots.
  storage       Version retention holds and observation periods; physical...
  contract      View and validate registered dataset data contracts.
  profile       View versioned research universe profiles.
  stats         Lake metrics under meta/stats (row counts, bytes, source mix).
  sources       Probe the data sources this lake depends on, and check the...
  delisted      Read the delisted directory; history is backfilled with
                backfill...
  ths-official  Cross-check and backfill against the THS (同花顺) official API...

安装方式: uvx
包名: cnequity
解析依据: pypi package lookup
沙箱: docker
总耗时: 45915 ms
安装耗时: 35666 ms
工作目录: 654 MB
探针版本: 1.0.0
实测时间: 2026-11-11 07:52:44

如何使用

❌ 实测失败runtime_error · Inspect historical decision-data evidence and unresolved；下面配置仅供参考，跑不跑得通取决于你的环境

1安装 / 启动命令uv / uvx

uvx cnequity

2客户端配置Claude Desktop / Cursor

{
  "mcpServers": {
    "rootsunc-cnequity": {
      "command": "uvx",
      "args": [
        "cnequity"
      ]
    }
  }
}

整段粘进 claude_desktop_config.json 的 mcpServers 里（Cursor 用 ~/.cursor/mcp.json），保存后重启客户端。

安装方式来源：pypi package lookup，沙箱实测未通过。

打开 GitHub 仓库 ↗项目主页 ↗MCP 体检工具

仓库信息

全名: rootSunc/CNEquity
分类: server
Stars: 997
Forks: 109
主要语言: Python
License: Apache-2.0
最近推送: 2026-11-06
仓库创建: 2026-06-28
首次收录: 2026-11-10
最近抓取: 2026-11-10

#a-share#a-shares#a-stock#ai-agent#akshare#ashare#astock#backtesting#baostock#china-stock-market#data-lake#financial-data