jin-duo-duo-strategy

基于技术分析的股票交易策略分析工具；当用户需要进行股票买卖点分析、技术形态识别、交易策略评估时使用

Security Scan

The skill's requirements, instructions, and included Python script align with a technical‑analysis trading helper; nothing in the package asks for unrelated credentials, network access, or elevated persistence.

Purpose & Capability

Name/description (technical stock-strategy analysis) match the delivered artifacts: SKILL.md, a strategy guide, and a Python script that computes MA/MACD/volume and performs signal detection. Required resources (pandas/numpy) are appropriate for the stated purpose.

Instruction Scope

Runtime instructions ask the agent to accept user-provided CSV/JSON OHLCV data, run the included script, read the local references/strategy-guide.md for rule interpretation, and produce a report — all consistent with the stated analysis task. There are no instructions to read unrelated system files, environment variables, or to send data to external endpoints.

Install Mechanism

No install spec is provided (instruction-only). The package includes a local Python script and a requirements.txt listing only pandas and numpy — proportionate and expected. No remote downloads, installers, or archives are referenced.

Credentials

The skill declares no required environment variables, credentials, or special config paths. That is proportionate for a local data-analysis tool that only processes user-supplied price/volume data.

Persistence & Privilege

The package metadata has always: false and no install-time persistence. One minor inconsistency: SOUL.md states the agent 'records user history' for tracking, but there is no code in the included script that writes or manages persistent storage (no file/database/network sinks shown). If the author intends history recording, ask where/how data is stored before enabling persistence.

Assessment

This package appears coherent and limited to local technical analysis. Before installing or running it: 1) Inspect the full scripts/technical_indicators.py file locally (search for network I/O, file writes, or unexpected subprocess calls); 2) Run it in a sandbox or non-production environment and test with the provided sample_data.csv; 3) Install only the declared dependencies (pandas, numpy) in a virtualenv; 4) If you expect the agent to 'remember' past requests (SOUL.md mentions recording history), confirm where those histories are stored and whether they leak sensitive data; 5) Never feed the skill private credentials or unrelated files — it does not require them and they are unnecessary for its stated function.

金多多股票策略分析 Skill

任务目标

• 本 Skill 用于：根据技术指标和K线形态分析股票，匹配6种交易策略
• 能力包含：技术指标计算、形态识别辅助、策略评分与推荐
• 触发条件：用户提供股票数据或询问买卖时机、策略分析时

前置准备

• 依赖说明：scripts 脚本所需的依赖包

  pandas>=2.0.0
  numpy>=1.24.0
  

• 输入数据准备：准备股票历史数据，包含以下字段
  • 日期、开盘价、最高价、最低价、收盘价、成交量
  • 支持 CSV 文件或 JSON 格式
  • 数据格式说明见 references/strategy-guide.md

操作步骤

• 标准流程：

  1. 数据准备：用户提供股票历史数据（CSV 文件或 JSON 格式）
  2. 指标计算：调用 scripts/technical_indicators.py 计算技术指标
     • 该脚本会计算移动平均线（MA5/10/20/60）、MACD 指标、成交量分析
     • 输出结构化的指标数据
  3. 形态识别与策略匹配：智能体根据计算结果和 K 线数据，识别技术形态
     • 识别 K 线形态（出水芙蓉、金针探底、看跌吞没等）
     • 匹配策略条件（见 references/strategy-guide.md）
  4. 评分与推荐：根据策略条件计算得分，输出匹配的策略和操作建议
  5. 生成报告：输出包含策略类型、操作建议、关键信号、风险提示的分析报告

• 可选分支：

  • 当 数据不足（少于20个交易日）→ 提示用户补充数据，策略判断可能不准确
  • 当 信号模糊（多个策略条件接近）→ 标记为"观望等待"（策略六），建议继续观察
  • 当 满足多个买入策略 → 选择得分最高的策略，并列出次要信号

资源索引

• 必要脚本：见 scripts/technical_indicators.py（用途：计算 MA、MACD、成交量等技术指标）
• 领域参考：见 references/strategy-guide.md（何时读取：策略匹配、形态识别时）
• 输出资产：无固定模板，根据分析结果动态生成报告

注意事项

• 本 Skill 仅基于技术分析，不构成投资建议
• 所有技术指标计算由脚本完成，确保数据准确性
• K 线形态识别和策略判断由智能体基于数据分析完成
• 评分机制仅供参考，实际操作需结合市场环境和个人风险偏好
• 建议至少提供 60 个交易日的数据以获得更准确的分析结果

使用示例

示例1：从 CSV 文件分析股票

• 功能说明：读取股票 CSV 数据，计算指标并匹配策略
• 执行方式：调用脚本 + 智能体分析
• 关键步骤：

  python scripts/technical_indicators.py --input stock_data.csv
  

• 指导要点：CSV 需包含日期、开盘、最高、最低、收盘、成交量字段

示例2：从 JSON 数据快速分析

• 功能说明：直接传入 JSON 格式的股票数据进行分析
• 执行方式：调用脚本 + 智能体分析
• 关键步骤：

  echo '{"data":[{"date":"2024-01-01","open":10.0,"high":10.5,"low":9.8,"close":10.2,"volume":1000000}]}' | \
  python scripts/technical_indicators.py --input-format json
  

• 指导要点：JSON 数组需包含完整的 OHLCV 数据

示例3：策略条件查询

• 功能说明：用户询问特定策略的判断条件
• 执行方式：仅智能体（无需脚本）
• 关键步骤：读取 references/strategy-guide.md 中的策略定义，解释给用户
• 指导要点：策略六（观望等待）是所有其他策略不匹配时的默认选项