← Back to Skills Marketplace
A股雷达
by
Bear Xiong
· GitHub ↗
· v1.1.0
· MIT-0
91
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install a-stock-radar
Description
A股综合监控与分析工具。当用户询问「A股今天怎么样」「大盘如何」「哪些板块在动」「今日涨跌」「北向资金」「涨停/跌停」「市场情绪」「龙虎榜」「现在适合上仓位吗」「分析A股」「A股复盘」「当前主线是什么」「A股核心企业」「茅台/宁德时代/比亚迪」「SHIBOR」「LPR」「中美10Y利差」「社融/人民币贷款」「A股...
README (SKILL.md)
A股雷达 (A-Stock Radar) v2.0
⚠️ 本工具输出包含寒鸦框架决策建议，但所有建议仅供研究参考，不构成任何投资建议。市场有风险，投资需谨慎。
🆕 v2.0 重大升级：寒鸦决策框架
除原有行情监控外，新增交易决策矩阵输出：
寒鸦四维框架
| 维度 | 内容 |
|---|---|
| 宏观定调 | LPR周期、SHIBOR流动性、中美利差、政策发令枪指向 |
| 微观验证 | 机构龙虎榜资金、筹码高低切、ETF异动 |
| 情绪水位 | 涨停/跌停/炸板率/连板高度 → 冰点/分歧/修复/亢奋 |
| 风控矩阵 | 仓位建议、买入板块、卖出板块、量化止损阈值 |
情绪数据修复（2026-04-29）
⚠️ 已知Bug：akshare的stock_zt_pool_em等函数默认日期为2024年10月，导致涨停数据永远为0。已在sentiment_snapshot.py中修复——所有API调用现已传入当日日期（YYYYMMDD格式）。
a-stock-radar 现在是统一的 A 股入口，内部按四层能力组织：
实时行情监控宏观快照与核心企业寒鸦决策框架（含情绪判断）
a-stock-radar 现在是统一的 A 股入口，内部按三层能力组织：
实时行情监控宏观快照与核心企业短线情绪判断深度量化分析
1. 实时行情监控
适用问题：
- A股今天怎么样
- 哪些板块在动
- 北向资金如何
- 涨停跌停情况
- 某只股票现在多少钱
主要脚本：
scripts/index_spot.py：主要指数scripts/sector_ranking.py：板块排行scripts/stock_quote.py：个股行情scripts/zt_pool.py：涨停池/跌停池scripts/dashboard.py：综合看盘入口scripts/macro_snapshot.py：SHIBOR、LPR、中美10Y利差、北向、新增人民币贷款scripts/core_companies.py：A股核心企业篮子
主要数据源：
- 东方财富：板块/排行
- 新浪财经：指数与个股实时行情
- akshare：涨跌停池、北向等扩展数据
2. 宏观快照与核心企业
适用问题：
- A股宏观数据怎么看
- SHIBOR/LPR 现在是什么水平
- 中美10Y利差如何
- 北向资金和融资需求怎么样
- 茅台/宁德时代/比亚迪现在怎么样
- 给我看A股核心企业篮子
主要脚本：
scripts/macro_snapshot.pyscripts/core_companies.py
覆盖维度：
- SHIBOR（银行间流动性）
- LPR（贷款市场报价利率）
- 中美10Y国债利差
- 北向资金净流入
- 新增人民币贷款
- 贵州茅台、宁德时代、比亚迪、招商银行、工商银行、美的集团、中芯国际、立讯精密
3. 短线情绪判断
适用问题：
- 今天市场情绪怎么样
- 现在是冰点还是亢奋
- 适不适合上仓位
- 短线环境强不强
主要脚本：
scripts/sentiment_snapshot.py
核心指标：
- 涨停家数
- 跌停家数
- 炸板率
- 连板高度
输出内容：
- 情绪阶段：
冰点 / 分歧 / 修复 / 亢奋 - 仓位建议：
空仓 / 轻仓 / 半仓 / 重仓 - 短线打法建议
4. 深度量化分析
适用问题：
- 分析A股
- 做一份A股复盘
- 当前主线是什么
- 用量化视角看A股
主要脚本：
scripts/quant_analysis.py
分析框架：
- 宏观流动性
- 政策主线
- ETF/龙虎榜/换手结构
- 情绪与筹码共振
输出内容：
- 主线判断
- 择时建议
- 复盘框架
- 风险提示
使用原则
- 只问盘面与行情：优先走
实时行情监控 - 问仓位与短线环境：优先走
短线情绪判断 - 问中高层判断与复盘：优先走
深度量化分析
迁移说明
以下旧 skill 已并入本 skill：
a-stock-market-sentimentashare-quant
Usage Guidance
This skill appears to do what it says (fetching A‑share market and macro data). Before installing/running: (1) run it in an isolated Python environment (venv/container) and install required packages from PyPI (requests, akshare, pandas, beautifulsoup4, lxml) so missing dependencies don't cause unexpected behavior; (2) review and, if needed, restrict network egress because the scripts fetch data from public finance sites (hq.sinajs.cn, push2.eastmoney.com) and akshare may query other sources; (3) be aware of rate limits and the legality/terms of scraping the target sites; (4) if you require offline/air‑gapped operation or stricter privacy, do not grant external network access; (5) if you want clearer safety, ask the author/maintainer to provide an explicit requirements.txt and an install/usage section in SKILL.md. Overall there are no hidden endpoints or credential exfiltration signs.
Capability Tags
Capability Assessment
Purpose & Capability
The name/description (A股监控与分析) matches the code: scripts fetch real‑time quotes, sector rankings, macro snapshots,龙虎榜 and sentiment. The network endpoints used (hq.sinajs.cn, push2.eastmoney.com, akshare-backed sources) are appropriate for the stated functionality.
Instruction Scope
SKILL.md describes routing to the included scripts and the scripts only perform web requests and local data processing; they do not read arbitrary system files, request environment secrets, or post data to unknown third‑party endpoints. All data flows appear limited to fetching market data from known public endpoints or via the akshare library.
Install Mechanism
There is no install spec (lowest install risk), but the package contains Python scripts that require third‑party libraries (requests, akshare, pandas, beautifulsoup4, lxml). These requirements are not declared in SKILL.md/registry metadata. Some scripts (e.g., zt_pool.py) will exit if akshare is missing. You should install these Python dependencies from trusted sources or run in an isolated environment.
Credentials
The skill requests no environment variables, no credentials, and no config paths. That is proportionate for a read‑only market data tool. Note: akshare itself may call various upstream data providers depending on implementation—this is expected but worth reviewing if you need strict network controls.
Persistence & Privilege
always is false and the skill does not modify other skills or system configuration. It runs as standalone scripts and spawns subprocesses only to invoke its own modules (dashboard uses subprocess to call local scripts). No elevated privileges are requested.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install a-stock-radar - After installation, invoke the skill by name or use
/a-stock-radar - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
feat: 新增龙虎榜数据模块；fix: 北向资金区分盘中未结算与数据缺失；情绪数据失败时不输出结论；板块错误信息细分原因
v1.0.0
首发：含宏观快照、核心企业、板块排行、情绪指标
Metadata
Frequently Asked Questions
What is A股雷达?
A股综合监控与分析工具。当用户询问「A股今天怎么样」「大盘如何」「哪些板块在动」「今日涨跌」「北向资金」「涨停/跌停」「市场情绪」「龙虎榜」「现在适合上仓位吗」「分析A股」「A股复盘」「当前主线是什么」「A股核心企业」「茅台/宁德时代/比亚迪」「SHIBOR」「LPR」「中美10Y利差」「社融/人民币贷款」「A股... It is an AI Agent Skill for Claude Code / OpenClaw, with 91 downloads so far.
How do I install A股雷达?
Run "/install a-stock-radar" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is A股雷达 free?
Yes, A股雷达 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does A股雷达 support?
A股雷达 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created A股雷达?
It is built and maintained by Bear Xiong (@gold3bear); the current version is v1.1.0.
More Skills