A股和港股股票分析助手

v1.0.10

股票监控分析技能 - 自定义股票池监控、实时行情、技术指标分析、涨跌趋势预测、信号提醒

1· 254·0 current·0 all-time
byLeon Xue@xsqorange

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for xsqorange/stock-monitor-orange.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "A股和港股股票分析助手" (xsqorange/stock-monitor-orange) from ClawHub.
Skill page: https://clawhub.ai/xsqorange/stock-monitor-orange
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install stock-monitor-orange

ClawHub CLI

Package manager switcher

npx clawhub@latest install stock-monitor-orange
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description match the code and instructions: scripts implement realtime quotes, K‑lines, technical indicators, capital flow and news aggregation using Tencent/EastMoney/Sina endpoints. Declared needs (python, network, local config files under ~/.openclaw) are consistent with the stated purpose.
Instruction Scope
Runtime instructions explicitly require reading/writing local files (~/.openclaw/stock-pool.json, stock-positions.json, stock-trades.json, stock-alerts.json, cron/jobs.json) and fetching data from several public financial APIs. Reading personal holdings/transaction records is expected for a portfolio monitor but is sensitive — the skill will access this personal data. SKILL.md also mentions pushing reports to Feishu/using agent message tools; the skill does not declare or implement Feishu credentials, so pushing likely relies on the host agent’s messaging integration.
Install Mechanism
No install spec; the skill is instruction + Python scripts that run with the system Python and standard library urllib. No downloads from external unknown URLs or package installs were specified.
Credentials
The skill does not request environment variables or external credentials. It does require access to local configuration files that contain personal holdings and trade history (expected for the feature but sensitive). It references pushing reports to Feishu without declaring credentials — this is probably intended to use the agent's built‑in messaging rather than external tokens, but users should confirm how report delivery is wired in their agent environment.
Persistence & Privilege
No elevated privileges requested: always:false, no system-wide configuration changes are declared, and the skill does not claim to persistently install daemons. It reads/writes files under the user's ~/.openclaw directory only (its own config scope).
Assessment
This skill is internally coherent for portfolio monitoring: it fetches market and news data from public Chinese finance APIs and reads your local portfolio/trade files under ~/.openclaw to compute signals and generate reports. Before installing, consider: (1) these local files contain sensitive personal financial data — back them up and review their contents; (2) the skill will make outbound HTTP(S) requests to public endpoints (qt.gtimg.cn, eastmoney, sina, etc.); if you have network restrictions or privacy concerns, review those endpoints; (3) report delivery mentions Feishu but no Feishu token is declared — verify how your agent will send reports (agent messaging vs. embedding credentials); (4) the included Python scripts contain some minor bugs/truncations in the distributed copy (e.g., incomplete/typoed returns and truncated prints), so test in a safe environment before using with real money or automating trades; (5) run the skill in a restricted environment or inspect/execute the scripts locally first if you are concerned about data leakage.

Like a lobster shell, security has layers — review code before you run it.

latestvk9744rvq2zzk3zrexf0c8czb4s84va2s
254downloads
1stars
13versions
Updated 1w ago
v1.0.10
MIT-0

Metadata

openclaw:
  requires:
    files:
      - ~/.openclaw/stock-pool.json        # 股票监控池配置(公开数据)
      - ~/.openclaw/stock-positions.json   # 持仓记录(个人数据,仅本地使用)
      - ~/.openclaw/stock-trades.json      # 交易记录(个人数据,仅本地使用)
      - ~/.openclaw/stock-alerts.json      # 预警配置
      - ~/.openclaw/cron/jobs.json         # 定时任务配置
    apis:
      - 腾讯行情API (qt.gtimg.cn)         # 实时行情(主要数据源)
      - 东方财富API (eastmoney.com)        # K线数据、资讯
      - 新浪财经API (sina.com.cn)          # 港股资讯
    network: true
    python: true

Stock Monitor - 股票监控分析技能


核心特色

  • 多维度技术分析:ADX趋势、BOLL轨道、MACD/KDJ/RSI/OBV/DMI/WR指标
  • 资讯情感分析:东方财富+新浪财经+雪球
  • 持仓智能管理:自动追踪成本、盈亏、FIFO同步
  • 买卖点位建议:基于支撑/阻力位+BOLL+均线的精确入场/出场/止损/目标价位
  • 中国股市习惯:🔴红色=上涨/盈利 | 🟢绿色=下跌/亏损

目录结构

stock-monitor/
├── SKILL.md                    本文件
├── references/
│   ├── commands.md             CLI命令速查
│   ├── config.md               配置文件说明
│   ├── index.md                指标说明
│   ├── scheduled-tasks.md      定时任务配置
│   └── troubleshooting.md      故障排查
├── reports/
│   ├── prompts.md              Cron任务Prompt模板
│   └── templates.md            报告模板(完整模板)
└── scripts/
    ├── stock_monitor.py        主监控脚本
    └── stock_capital.py        资金流向脚本

CLI 命令

工作目录:scripts | 格式:python stock_monitor.py <command> [args]

行情与指数

命令说明
quote <code>查询单只股票行情
index查询大盘指数

技术分析

命令说明
analyze <code>分析股票技术指标
monitor-a监控所有A股
monitor-hk监控所有港股

综合报告

命令说明
report生成完整综合报告
report-a生成A股综合报告
report-hk生成港股综合报告

持仓管理

命令说明
position add <code> <qty> <cost>添加/更新持仓
position remove <code>清除持仓
position list查看持仓

交易记录

命令说明
trade buy <code> <qty> <price>记录买入
trade sell <code> <qty> <price>记录卖出
trades [code]查看交易记录

重要规则

数据优先级(强制)

  1. 脚本行情数据(index/monitor-a/monitor-hk)— 权威数据
  2. 搜索资讯数据(web_fetch)— 仅用于资讯分析

【强制】报告中所有"现价/涨跌幅/涨跌额"必须使用脚本返回的数据,禁止使用搜索结果中的价格。

技术指标(必须包含)

每只股票:现价、涨跌幅、MA均线多空、MACD状态、KDJ状态、RSI数值、BOLL位置、OBV背离、DMI趋势、WR威廉、综合信号

综合信号评级

评分差信号
buy > sell + 3⭐⭐⭐ 强烈买入
buy > sell⭐⭐ 建议买入
abs差 ≤ 3⚪ 观望
sell > buy⭐ 建议卖出
sell > buy + 3⭐⭐ 强烈卖出

买卖点位计算规则

场景买入点位买入区间卖出点位卖出区间止损位目标位
回踩支撑支撑位支撑×0.98~1.02--支撑下方3%阻力位/BOLL上轨
突破阻力突破后确认阻力~阻力×1.02--阻力下方3%上一阻力位
RSI超卖BOLL下轨/支撑下轨×0.98~1.02--前低下方3%BOLL上轨/阻力位
RSI超买--BOLL上轨/阻力位上轨×0.97~1.01--
MACD死叉--短线阻力位阻力×0.97~1.01--

计算依据

  • 支撑位:近30日低点、BOLL下轨、60日均线
  • 阻力位:近30日高点、BOLL上轨、近期反弹高点
  • 止损:支撑/均线下方3%(保守可设2%)
  • 目标位:阻力位、BOLL上轨、量度涨幅(前一高低点差)
  • 买入区间:在点位附近±2%范围分批建仓
  • 卖出区间:在点位附近±3%范围分批止盈

报告模板

详见 reports/templates.md

报告生成流程

  1. python stock_monitor.py index 获取大盘指数
  2. 读取 ~/.openclaw/stock-pool.json 获取股票列表
  3. python stock_monitor.py monitor-a/hk 获取技术面数据
  4. web_fetch 获取资讯(优先东方财富/新浪/雪球)
  5. 生成报告(价格数据必须用步骤3返回的数据)
  6. 推送至飞书群聊

配置文件

文件路径
股票池~/.openclaw/stock-pool.json
持仓记录~/.openclaw/stock-positions.json
交易记录~/.openclaw/stock-trades.json
预警配置~/.openclaw/stock-alerts.json
Cron任务~/.openclaw/cron/jobs.json

使用提示

  • 技术指标有滞后性,用于确认趋势而非预测
  • 多条件共振更可靠,单一指标容易假信号
  • 动态止盈:回撤5%减仓、10%清仓(建议,根据市场灵活调整)
  • 核心原则:预警系统目标是"不错过大机会,不犯大错误"

Comments

Loading comments...