老谷拆财报 MCP Server(laogu-mcp)
16 个「老谷拆财报」财经 skill 的程序化数据层:A 股行情、公告、龙虎榜、两融、 解禁、估值……一次装好,16 个场景 tool 随便调。
零 key、零登录、零配置:全部走公开接口,不用申请任何 token。 以数据为刃,剖市场真相 —— 个人观点,仅供参考,不构成投资建议。
30 秒装好(小白版)
- 装 Python 3.10 或更新(python.org 下载,一路"下一步")。
- 装 uv(命令行工具,复制粘贴下面一行):
- Windows(PowerShell):
powershell -c "irm https://astral.sh/uv/install.ps1 | iex" - Mac/Linux:
curl -LsSf https://astral.sh/uv/install.sh | sh
- Windows(PowerShell):
- 测试一行能跑:
uvx laogu-mcp(看到启动日志即成功,按 Ctrl+C 退出)。
一键安装
uvx laogu-mcp
Claude Desktop 配置
打开 Claude Desktop → 设置 → 开发者 → MCP 服务器 → 编辑配置文件,加入:
{
"mcpServers": {
"laogu-mcp": {
"command": "uvx",
"args": ["laogu-mcp"]
}
}
}
保存后重启 Claude Desktop,对话框输入框下方出现 🔌 图标即装好。
Cherry Studio 配置
设置 → MCP 服务器 → 添加服务器 → 类型选"标准输入输出 (stdio)":
- 名称:
laogu-mcp - 命令:
uvx - 参数:
laogu-mcp
保存后开关打开,状态显示"已连接"即成功。
16 个 tool 一览
| tool | 对应 skill | 一句话 |
|---|---|---|
quote |
laogu-fundamentals | 个股行情快照(现价/涨跌/成交量额) |
ann_list |
laogu-announcements | 个股公告列表(含取正文的 art_code) |
code_verify |
laogu-news | 股票代码↔官方简称核对 |
market_snapshot |
laogu-morning | A股指数+美股+原油+黄金+离岸人民币 |
fund_flow |
laogu-moneyflow | 个股两融 / 全市场龙虎榜资金 Top |
close_recap |
laogu-close | 收盘指数快照 |
lhb_board |
laogu-lhb | 龙虎榜明细(自动回滚到最近交易日) |
ann_content |
laogu-report | 公告正文全文抓取(自动翻页) |
risk_inputs |
laogu-risk | 财务风险体检输入(行情+定期报告定位) |
earnings_ann |
laogu-earnings | 业绩预告/快报公告扫描 |
research_grounding |
laogu-research | 研报解读 grounding(代码核对+行情) |
ir_records |
laogu-notes | 投资者关系活动记录表搜索 |
unlock_notices |
laogu-unlock | 限售解禁公告搜索 |
ipo_calendar |
laogu-ipo | 打新日历(无公开接口时诚实返回搜索模板) |
valuation |
laogu-value | 估值锚输入(PE/PB/市值) |
macro_helper |
laogu-macro | 交易日校验+北京时间换算+宏观搜索模板 |
返回格式
统一 JSON 信封,成功失败都一样的结构:
{
"ok": true,
"data": { "...": "..." },
"meta": {"source": "新浪行情", "fetched_at": "2026-09-29 03:55:00", "timezone": "北京时间"},
"warnings": ["腾讯行情不可用,已降级到新浪行情"]
}
失败时 ok: false,带 reason 和 search_templates(网页搜索关键词),绝不编数字。
取不到的字段标 null 并在 warnings 说明,不估算。
诚实纪律(和 16 个 skill 同标准)
- 关键数据带日期时点和来源;单源数据标注来源。
- 北向"净流入/净买入"口径已死(2024-08-19 起),本 server 不输出该数字。
- 只做数据抓取,解读由宿主 LLM 按各 skill 的 Output Contract 执行;不做买卖推荐。
- 公共接口不提供具体龙虎榜营业部,绝不编造。
数据源实测状态(2026-09-29)
| 接口 | 状态 |
|---|---|
| 新浪行情(含指数/美股/期货/外汇) | ✅ |
| 东财公告列表 / 正文 | ✅ |
| 东财龙虎榜 / 两融 | ✅ |
| 腾讯行情 | ⚠️ 部分网络超时(自动降级) |
| 东财 push2 | ⚠️ 部分网络 502(自动降级) |
开发
git clone https://github.com/laogu-caibao/laogu-mcp
cd laogu-mcp
uv venv && uv pip install -e ".[dev]"
python smoke/run.py # 冒烟测试(真实数据)
—— 老谷拆财报 · laogu-mcp · 出品:老谷拆财报(抖音/视频号/头条/快手同名) 个人观点,仅供参考,不构成投资建议
mcp-name: io.github.laogu-caibao/laogu-mcp
Release files for laogu-mcp 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| laogu_mcp-1.0.1.tar.gz | 7.2 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| laogu_mcp-1.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 7.2 MB
Release files / laogu_mcp-1.0.1.tar.gz
| Download URL | laogu_mcp-1.0.1.tar.gz |
|---|---|
| Size | 7.2 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
c694be5e9feab3d28fe6e1d6380167f7cdae392f1356b0ae12b1e991140674aa
|
|
BLAKE2b-256 checksum How to use checksums |
798028f4c5ef194ece80a1f218ed645f96b5dc1c0e555e53c4ce4648b0389519
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 29, 2026.
Transparency logRelease files / laogu_mcp-1.0.1-py3-none-any.whl
| Download URL | laogu_mcp-1.0.1-py3-none-any.whl |
|---|---|
| Size | 16.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7cca84f16054f108cbffd337d478f459ad4fe520ee107bf1cecc784a680c33f3
|
|
BLAKE2b-256 checksum How to use checksums |
9c7cceac1a9c7108e478a89dad6c1ff5496d3d64a1160cdcd410b025b87d83f2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 29, 2026.
Transparency log