DayTrading.Monster

View saved version →

---
name: cn-market-tape
description: Use when the user asks for A-share intraday or after-close market tape, theme strength TOP10/BOTTOM10, theme/concept-board money-flow rankings, explicit industry-board rankings, limit-up pool, institutional survey heat, or a combined read of these signals. Treat an unqualified “板块” as a theme/concept board by default; use an industry board only when explicitly requested.
---

# CN Market Tape

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

统一处理 A 股盘中和盘后盘面数据。默认只抓用户请求的模块，不读取或修改用户自选股之外的账户数据，不把一日榜单、资金流和机构调研混成一个结论。

## Routing

先判断用户要的是盘中快照、收盘数据还是历史数据，再选择模块：

1. `题材强弱`、`题材 TOP10/BOTTOM10`：使用本 skill 的加权题材流程。
2. `板块流入流出`、`主力净流入/净流出`：默认按题材/概念板块处理，并直接使用 `push2delay.eastmoney.com` 聚合公开接口；该接口失败后只切换一次到 `push2.eastmoney.com`。只有用户明确说“行业板块/行业”时才切换到行业板块口径。`mx-data` 仅用于用户明确要求 MX 或公开聚合接口缺少的补充字段，不再作为该模块的前置调用。
3. `涨停池`：先尝试 MX 或已有的聚合涨停池接口；不要逐只股票抓取涨停状态。
4. `机构调研`：当前/最近交易日优先运行本 skill 内置的机构调研聚合脚本或 MX 查询；历史请求优先使用公开历史调研热度数据，找不到对应日期或字段时明确声明不支持。

若用户只要求“盘中快照”，不要自动抓股吧、新闻或历史数据；若用户要求“为什么”或“复盘”，再按需要调用 `mx-search`、`cn-stock-move-reason` 和 `macro-news-check`。

### DTM 接口

DTM 数据优先使用 `https://daytrading.monster/api-docs/` 中的正式接口：

- `https://daytrading.monster/api/chinastock-anomaly`：最新涨跌题材 Top10、涨停概念 Top8、当日小时榜单、完整异动原因和涨停池。逐块核对交易日和时间；该榜单与下文 MX 成分加权榜是不同口径，不能混合排名或当作同口径历史快照。
- `https://daytrading.monster/api/themes?market=cn`：A 股题材成员、权重、中文名称与依据及阶段表现；阶段涨跌与 `price_dates` 不作为实时行情。此接口用于生成映射缓存；加权题材榜继续读取 MX 行情，资金榜与分时按 Module 2 的来源规则执行。

接口正文是 JSON，即使 Content-Type 为 `text/plain`。不读取 DTM 机构调研数据；机构调研继续使用既有流程。

## Dependencies

题材强弱的必需依赖：

- `mx-zixuan`：先取一次用户自选股行情，作为快速覆盖层。
- `mx-xuangu`：按代码批量补抓题材映射中未覆盖的 A 股成分。
- `mx-search`：默认只对 TOP3 题材做驱动核查。

其他模块：

- `mx-data`：用于涨停池、机构调研和指定字段的补充查询；板块资金榜及分时不再默认前置调用。
- `visualize`：用户要求分钟级折线图时，用于渲染带零轴和拐点标注的分时资金图；不可用时退回分钟数据表并说明限制。
- `scripts/institutional_survey_heat.py`：机构调研明细的低频抓取、去重和股票/行业/周度聚合。
- `cn-stock-move-reason`：只有用户需要解释涨停池或个别板块异动时再调用。
- `macro-news-check`：只有用户要求把宏观、跨市场或大盘因素纳入时再调用。

MX 未配置、认证失败、超 quota 或返回空数据时，不要假装已有结果。按 `references/market-tape-source-routing.md` 切换到明确的聚合备用源，并在状态行报告口径变化。

## Module 1: Theme Strength

### Theme universe

题材强弱榜使用本地 A 股题材映射缓存；刷新脚本从正式 `/api/themes?market=cn` 的 `themes[].constituents[]` 提取成员和权重，并从 `theme_name_zh` 生成中文标签。缓存文件为：

- `assets/themes/theme-data.json`
- `assets/themes/theme-label-i18n.json`

正常运行时执行一次映射刷新脚本；两份缓存都有效且未超过 7 天时跳过网络刷新：

```bash
python3 <cn-market-tape>/scripts/refresh_theme_assets_from_public_source.py
```

刷新失败但缓存有效时继续运行，并在状态行写明“题材映射刷新跳过”。缓存缺失或无效且刷新失败时停止题材强弱模块，不得从 MX 行业名临时发明题材池。

只保留 `market == "CN"`、`theme != "未分類"` 的行。每行至少需要 `code`、`theme` 和 `weight`；重复的股票-题材关系按独立权重参与计算。中文标签优先取 `theme-label-i18n.json[theme].zh`。

### Quotes and aggregation

1. 提取去重后的 A 股代码，但保留所有股票-题材关系用于加权。
2. 调用 `mx-zixuan` 一次，读取 `SECURITY_CODE`、`SECURITY_SHORT_NAME`、`NEWEST_PRICE`、`CHG` 和 `dateMsg`。
3. 对缺失代码按最多 50 只一批调用 `mx-xuangu`，批次串行。遇到 `操作过于频繁` 或 HTTP 503，等待 8-15 秒后重试同一批，最多重试 2 次。
4. 合并行情时优先使用 `mx-zixuan`，用 `mx-xuangu` 补齐；若日期不一致，报告混合日期风险。
5. 按下式计算每个题材的加权涨跌幅：

```text
theme_return = sum(theme_weight * stock_chg_pct) / sum(theme_weight)
```

`CHG` 按百分数处理，例如 `10.03` 表示 `+10.03%`。TOP 榜列出 `weight * CHG` 最大的贡献股，BOTTOM 榜列出最小的拖累股。

### Theme interpretation

需要判断持续性、拥挤度、主线、反抽或退潮时，读取 `references/theme-mainline-lifecycle.md`。生命周期只解释榜单质量，不替代加权排名。

默认只检查 TOP3 题材：每个题材选涨幅最高的代表股，结合股吧/讨论发现和 `mx-search` 资讯，区分确认消息、市场思惑、海外映射和个股独立逻辑，并给出“较高/中等/较低”确定度。

## Module 2: Theme/Board Money Flow

### Source priority

先确定板块命名空间：用户说“板块/题材/概念”时默认使用题材/概念板块；只有明确说“行业板块/行业”时才使用行业板块。随后直接请求 `push2delay.eastmoney.com` 聚合公开接口，并校验目标板块名称、板块级资金字段和目标交易日；默认接口失败后只切换一次到 `push2.eastmoney.com`。`mx-data` 仅用于用户明确要求 MX 或聚合接口缺少的补充字段。不要逐板块、逐股票循环抓取。

公开备用接口的命名空间固定为：概念/题材板块 `m:90+t:3`；行业板块 `m:90+t:2`。每次输出必须注明“概念板块”或“行业板块”，不得把两个宇宙混在同一张榜或同一组快照比较中。

备用源的字段、请求顺序、超时和错误处理见 `references/market-tape-source-routing.md`。排名榜必须分别获取净流入方向和净流出方向；不能只取按降序返回的第一页，再把末尾几行误称为净流出。若接口声明的总数超过本次返回条数，要记录分页/返

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 10 — file references are metadata, not package contents.

View saved version →

---
name: cn-stock-move-reason
description: Use when analyzing why one A-share stock moved sharply from Eastmoney quote data, announcements, Eastmoney 股吧/资讯 posts, Eastmoney Guba topic heat, Sohu index/sector context, and A-share breadth.
---

# CN Stock Move Reason

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill when the user asks why one A-share stock is rising, falling, 涨停, 跌停, 炸板, 异动, or moving unusually.

The bundled script is safe for a public repository: it uses only public web pages/APIs, does not read credentials, and does not call any LLM service. Never commit personal information, API keys, account data, private RAG files, or private research materials to GitHub.

The skill accepts one stock at a time. It may also collect broad market context (indexes, sector/concept boards, and advance/decline counts) to judge whether the move is market-wide, sector-led, or stock-specific.

## Workflow


## Public Release

When preparing a GitHub upload or public release, use the public version only and run the release/privacy check in this Skill at `references/release-and-privacy.md`. Never upload private RAG folders, `.ftindex` files, credentials, `.env`, personal data, screenshots, raw PDFs/PPTs, or private strategy labels.


1. Read `references/experience.md` before analysis, but only the `Active Playbook` and `Compression Protocol` sections unless the user explicitly asks for historical lessons. Apply those lessons when judging catalysts, sector共振, 股吧 emotion, A-share emotion-cycle position, and the stock's place in the market structure: 主线、助攻、补涨、防御、老龙反抽, or noise. When the request needs a deeper or reusable sentiment framework, also use `stock-sentiment-analysis` and its `references/sentiment-framework.md`.

   Cross-skill calls are operational. When this workflow says to use another market skill, actually load that skill's `SKILL.md` and required references if the skill is installed or available as a sibling in this repository. Do not merely mention the other skill by name in the answer.

   Required coordination: for A-share analysis, use this skill as the evidence-gathering entry point, and add supporting skills based on clues found during analysis, not only on the user's wording. 股吧 itself is an evidence source, not an automatic trigger for sentiment analysis. If 股吧/资讯/news reveals a concrete clue about main-line status, crowding, attack/defense rotation, leader/follower position, or emotion-cycle phase, load `stock-sentiment-analysis` to test that clue. If 股吧/资讯/news discusses the broad market, index pressure, policy/liquidity, FX/rates, commodities, overseas markets, or geopolitical drivers, load `macro-news-check` to verify the tape instead of accepting forum claims. If 股吧/资讯/news or the price move points to support/resistance, failed breakout, trend damage, or catalyst acceptance/rejection, load `stock-technical-analysis` to verify the chart. When the original question is directly about an index/broad tape such as 上证、深成指、创业板、科创50、A50、恒生科技/港股 spillover, load `macro-news-check` by default.

   Mandatory execution gate:
   - If the answer uses **宏观** or **快讯** to explain the stock, sector, index, liquidity, policy, FX/rates, commodities, overseas spillover, or geopolitics, load `macro-news-check`. Do not replace this layer with ad hoc web search.
   - If the answer uses **技术面** such as support/resistance, trend confirmation, failed breakout, volume-price behavior, intraday timing, or "能不能上/下", load `stock-technical-analysis`.
   - If the answer uses **情绪面** such as main-line status, crowding, emotion cycle, leader/follower, risk-on/risk-off, expectation gap, or forum psychology, load `stock-sentiment-analysis`.
   - If the user asks about **个股-板块-大盘共振**, a collective sector surge, or whether a sector/stock move can continue, load `stock-sentiment-analysis` and ap

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 6 — file references are metadata, not package contents.

View saved version →

---
name: jp-stock-move-reason
description: Use when analyzing why a Japanese stock moved sharply from Yahoo Finance quote data, Yahoo 掲示板 comments, Yahoo/Kabutan/Traders news, and basic stock metrics such as current change, market cap, PER/PBR, dividend yield, and margin ratio.
---

# JP Stock Move Reason

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill from the installed package root when the user asks why a specific Japanese stock is rising, falling, 急騰, 急落, 異動, or otherwise moving.

The bundled script is safe to keep in a public repository: it uses only public web pages/APIs, does not read credentials, and does not call any LLM service. Never commit personal information, API keys, account data, private RAG files, or private research materials to GitHub.

## Workflow


## Public Release

When preparing a GitHub upload or public release, use the public version only and run the release/privacy check in this Skill at `references/release-and-privacy.md`. Never upload private RAG folders, `.ftindex` files, credentials, `.env`, personal data, screenshots, raw PDFs/PPTs, or private strategy labels.


1. Read `references/experience.md` before analysis, but only the `Active Playbook` and `Compression Protocol` sections unless the user explicitly asks for historical lessons. Apply those lessons when setting expectations, especially around earnings, guidance, valuation, 掲示板 sentiment, theme leadership, peer follow-through, and whether the stock is a leader, follower, defensive alternative, old-leader rebound, or noise. When the request needs a deeper or reusable sentiment framework, also use `stock-sentiment-analysis` and its `references/sentiment-framework.md`.

   Cross-skill calls are operational. When this workflow says to use another market skill, actually load that skill's `SKILL.md` and required references if the skill is installed or available as a sibling in this repository. Do not merely mention the other skill by name in the answer.

   Required coordination: for Japanese stock analysis, use this skill as the evidence-gathering entry point, and add supporting skills based on clues found during analysis, not only on the user's wording. Yahoo 掲示板 itself is an evidence source, not an automatic trigger for sentiment analysis. If 掲示板/news reveals a concrete clue about theme leadership, peer follow-through, crowding, leader/follower position, defensive alternative, old-leader rebound, or risk-on/risk-off acceptance, load `stock-sentiment-analysis` to test that clue. If 掲示板/news discusses Nikkei/TOPIX, JPX sectors, Nikkei futures, JGB yields, USD/JPY, BOJ/MOF policy, US/China spillover, commodities, or geopolitics, load `macro-news-check` to verify the tape instead of accepting forum claims. If 掲示板/news or the price move points to support/resistance, failed breakout, trend damage, or catalyst acceptance/rejection, load `stock-technical-analysis` to verify the chart. When the original question is directly about an index/broad tape such as Nikkei 225, TOPIX, JPX sectors, Nikkei futures, or 日经大盘, load `macro-news-check` by default.

   Mandatory execution gate:
   - If the answer uses **宏观** or **快讯** to explain the stock, sector, Nikkei/TOPIX, JGB yields, USD/JPY, BOJ/MOF policy, US/China spillover, commodities, or geopolitics, load `macro-news-check`. Do not replace this layer with ad hoc web search.
   - If the answer uses **技术面** such as support/resistance, trend confirmation, failed breakout, volume-price behavior, intraday timing, or "能不能上/下", load `stock-technical-analysis`.
   - If the answer uses **情绪面** such as theme leadership, crowding, leader/follower, defensive alternative, old-leader rebound, risk-on/risk-off, expectation gap, or 掲示板 psychology, load `stock-sentiment-analysis`.
   - Final answers should include a compact `融合口径` line when any supporting skill is 

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 10 — file references are metadata, not package contents.

View saved version →

---
name: macro-news-check
description: Use when stock, index, gamma, or market analysis needs current macro or broad-market context from live news sources such as Jin10, Wallstreetcn, FinancialJuice, and market tape confirmation sources such as Sohu for A-share sector/index breadth and JPX for Japanese real-time index/sector strength. Especially useful for rates, FX, central banks, commodities, geopolitics, index moves, market-wide risk sentiment, sudden cross-asset news, A-share broad-market/sector rotation questions, or Japan market breadth/sector-drag questions.
---

# Macro News Check

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill only when the analysis genuinely needs current macro, broad-market, or cross-asset context. Do not run it for every single-stock question by default.

This skill is public-safe: it uses public pages or feeds and must not store credentials, cookies, account data, private research paths, or raw copyrighted news dumps. Summarize only the headlines and implications needed for the user's market question.

## Trigger Conditions

Call this skill when one or more of these are true:

- The user asks about 大盘, 宏观, 利率, 汇率, 美债, 日债, 央行, CPI/PCE/FOMC/BOJ/ECB, commodities, oil, gold, geopolitics, or market-wide risk sentiment.
- A stock/index/ETF is moving with no clear stock-specific catalyst, or the move may be driven by rates, FX, futures, sector-wide risk, policy, or global headlines.
- Technical analysis needs a broad-market confirmation, especially for index breaks, high-beta stocks, export-sensitive Japanese names, A-share market emotion, US premarket moves, or same-day trading judgments.
- Gamma analysis would be unreliable without checking macro headlines, event risk, vol shock, index futures, yields, USD, or geopolitical tape risk.
- U.S. stock-move or gamma workflows such as `us-stock-move-reason` and `us-stock-gamma-moomoo` involve Fed expectations, rates, USD, oil/gold, index futures, market-wide risk appetite, or geopolitical shocks.
- A prior single-stock read may be missing a market-wide reason for a selloff or squeeze.

Do not call this skill when the request is clearly only about static fundamentals, historical valuation, company filings, or a chart level that does not depend on current market tone.

## Source Order

Prefer sources in this order, adjusting for language and market:

1. `Jin10` (`https://www.jin10.com/`): preferred Chinese macro tape when accessible. Its data quality is generally strong for real-time Chinese-language macro, central-bank, commodity, FX, and geopolitical headlines. The homepage can expose server-rendered flash items; the underlying flash API or WebSocket may require frontend headers/cookies and can be more fragile.
2. `Wallstreetcn` (`https://wallstreetcn.com/live/global`): useful Chinese backup and Asia/China market tape. If the page is a frontend shell, try its live JSON endpoint pattern:
   `https://api-one-wscn.awtmt.com/apiv1/content/lives?channel=global-channel&client=pc&limit=10`
3. `FinancialJuice` (`https://www.financialjuice.com/`): useful English global tape, especially US/EU macro, rates, FX, commodities, geopolitics, and market-moving headlines. Public RSS pattern:
   `https://www.financialjuice.com/feed.ashx?xy=rss`
   Tested shell fetch:
   `curl -L 'https://www.financialjuice.com/feed.ashx?xy=rss'`
   Quote the URL in zsh or other shells where `?` can be treated as a glob pattern. The feed returns broad global headlines; fetch the full feed first, then filter by macro relevance.
4. `DayTrading.monster 24H Feed` (`https://daytrading.monster/api-docs/`): usable as a public current-page social-media layer for global macro, AI, rates/bonds, US index/gamma, Japan, China, commodities, non-AI sectors, and crypto. For broad U.S. or Japan market context it is especially useful for theme discovery;

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 4 — file references are metadata, not package contents.

View saved version →

---
name: market-calendar-google
description: Organize a selected week of US earnings calendars or China/US/Japan macro and market-event calendars, prioritize what matters to the user, and add the resulting events to Google Calendar. Use when the user asks to handle this week's or next week's earnings, Earnings Whispers images, US stock ticker earnings, Treasury auctions, central-bank/data releases, or China/US/Japan financial events and wants them written to Google Calendar. Also use for rolling monthly expiry, settlement and Japan SQ calendar maintenance, and to view, save, revise, or remove local calendar preferences, including event-summary style.
---

# Market Calendar Google

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

## Overview

Use this skill to turn weekly market calendars into concise Google Calendar events for the user. Support four workflows:

1. US earnings calendar for a week, usually from the Earnings Whispers "Most Anticipated Earnings Releases" image.
2. China/US/Japan macro, central-bank, auction, and market-event calendar for a week.
3. Japan stock earnings calendar for a week, usually from SBI Securities settlement announcement data.
4. Rolling monthly expiry, settlement and Japan SQ date markers. For recurring calendar maintenance or requests about these dates, read [monthly market structure maintenance](references/monthly-market-structure.md) and apply it alongside the weekly workflow.

Default to the user's local timezone from the runtime environment. Use the current date and timezone from the environment to resolve "this week" and "next week". If the user's timezone is unavailable, ask for the target timezone before writing Calendar events.

## Local Preference Reference

- Keep two layers: public references contain shared calendar methods and ship with the Skill; a separate private reference contains each user's own preferences and stays on that user's machine. Maintain both; do not turn the public references into a personal preference file.
- When the user asks to remember a preference for future calendars, or to view, revise or remove one, use this local-reference workflow. Preference-only requests need filesystem access, not Google Calendar access. A one-off instruction applies to the current task without becoming a saved default.
- Read the private index through the reference contract above and reuse the calendar preference document it maps. If no mapping exists, create `calendar-preferences.md` beside that index and register it under `market-calendar-google`; when no index exists, initialize it at the contract's default location. Keep both outside the public repository and installed Skill/plugin directories. If local files are unavailable, provide the proposed text and report that it was not saved.
- Save only the user's stated preference, its scope and update date; name the public section it overrides when applicable. Supported preferences include summary length/structure, content to emphasize or omit, language, title style, event selection, destination calendar and reminders. Do not invent values for unspecified preferences or store temporary market forecasts as preferences.
- Update the matching entry in place, preserving unrelated preferences. A clear new permanent choice replaces its prior value; a temporary override leaves the saved default intact. Remove only the requested preference. Ask only when the intended scope or replacement is unclear.
- On calendar tasks, apply the current request first, then applicable saved preferences, then public defaults, subject to the reference contract's truthfulness and permission boundaries. Saving preferences does not itself create, migrate or rewrite Calendar events.
- Read back the changed entry and index mapping, then briefly confirm what was saved, revised or removed. Local preference content and priva

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 6 — file references are metadata, not package contents.

View saved version →

---
name: market-daily-strategist
description: Use when the user asks for daily market strategy, broad-market trading guidance, pre-market strategy, close recap, or daily/one-name long-term recommendation reports covering US stocks, Japanese stocks, or A-shares. Also use for natural current-session or next-session market questions such as "明天日股如何做", "今天日股怎么做", "日经明天怎么看", "美股今晚怎么做", "纳指今晚怎么看", "A股明天怎么操作", or "大盘今天怎么办" when the intent is market-level strategy. Applies to ad-hoc requests that need live market/news data, strict no-fabrication data discipline, price-confirmed catalysts, and decision-oriented trading guidance.
---

# Market Daily Strategist

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill for the user's market reports. This is not a scheduler; ignore any clock-trigger wording from the original prompts. Route by user intent:

- 美股盘前、开盘前、pre-market、盘前策略、美股今晚怎么做、美股明天如何做、今天美股怎么看、纳指今晚怎么看、标普明天怎么看、美国市场怎么做、美股大盘策略: read `references/us-pre-market.md`.
- 美股收盘、昨晚美股、复盘、close recap: read `references/us-close-briefing.md`.
- 日股盘前、日经盘前、日本开盘前、明天日股如何做、今天日股怎么做、日经明天怎么看、日本市场怎么做、日股大盘策略: read `references/jp-pre-market.md`.
- 日股收盘、今天日股复盘、日本市场复盘: read `references/jp-close-briefing.md`.
- A股盘前、A股早盘、开盘前策略、明天A股如何做、今天A股怎么做、A股明天怎么操作、大盘今天怎么办、沪指明天怎么看、创业板明天怎么看、科创板今天怎么做、A股大盘策略: read `references/cn-pre-market.md`.
- A股收盘、A股复盘、今天A股市场回顾: read `references/cn-close-briefing.md`.
- 美股长线推荐、每日美股荐股、推荐一只美股: read `references/us-long-term.md`.
- 日股长线推荐、每日日股荐股、推荐一只日股: read `references/jp-long-term.md`.
- A股长线推荐、每日A股荐股、推荐一只A股/ETF/LOF: read `references/cn-long-term.md`.

If the user asks a broad current-session or next-session question such as `明天/今天/今晚 + 市场/大盘/美股/纳指/标普/日股/日经/TOPIX/A股/沪指/创业板/科创板 + 如何做/怎么做/怎么看/怎么操作/怎么办`, treat it as a pre-market, current-session, or next-session strategy request for the named market, not as a generic macro question. If the market is named by an index, map it to the corresponding market reference above. If the user says only `盘前信息`, `收盘复盘`, `大盘怎么做`, or `推荐一只股票`, infer the market from the conversation. If unclear, ask one concise question for the market: 美股、日股、还是A股.

Always read `references/shared.md` first, then only the one task-specific reference that matches the user request.

For any daily or one-name recommendation, also read `references/strategy-archetypes.md`. Assign one primary strategy archetype before choosing entry, invalidation/stop, holding period, and exit. Do not combine an event-trade entry with a value-investing stop, or a swing entry with an indefinite long-term exit.

Reference reading rule: when this skill selects a reference file, first scan the file structure, then read the sections and nearby guardrails relevant to the current task. Read the complete file when it is short, when the task is broad or strategic, or when partial reading could miss constraints. Do not rely on stale memory or heading-only scans.

During question decomposition, actively expand broad market, sector, theme, and long-term stock questions into `mainline x funds x game x cycle` when that lens can change the conclusion, even if the user did not explicitly ask for it. In those cases, read the relevant sections of `references/global-mainline-funds-game-cycle.md`. When the conclusion depends on whether a collective sector/theme surge can continue, load `stock-sentiment-analysis` and apply its market-sector-stock resonance framework across A-shares, U.S. stocks, or Japanese stocks.

## Core Workflow

1. Identify the report type and market.
2. Confirm current date/time in Japan time and whether the relevant next or current session is open. If the market is closed, follow the task-specific closed-market rule instead of forcing a normal report.
3. Gather latest data from reliable live sources. Use market-specific primary sources listed in the task reference.
4. For U.S. and Japan market re

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 16 — file references are metadata, not package contents.

View saved version →

---
name: stock-sentiment-analysis
description: Reusable public-safe sentiment and market-emotion framework for A-shares, Japanese stocks, US stocks, indexes, and sector themes. Use when a stock or market move needs emotion-cycle classification, main-line versus follower judgment, expectation-gap analysis, forum/news sentiment synthesis, risk-on/risk-off context, market-sector-stock resonance, post-surge continuation analysis, A-share sector/theme constituent mapping such as 哪些股票/相关股/概念股/龙头股, or when other stock skills need a shared sentiment layer. Supports optional user-specified private RAG folders without storing or publishing private materials.
---

# Stock Sentiment Analysis

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill as the shared sentiment layer for market skills. It does not fetch data by itself; it tells Codex how to interpret evidence gathered by `cn-stock-move-reason`, `jp-stock-move-reason`, `stock-technical-analysis`, `us-stock-gamma-moomoo`, public news, forums, breadth, and user-provided screenshots or notes.

This public-safe skill must not contain personal information, API keys, account data, private paths, raw screenshots, full copied notes, ticker-specific personal trade logs, or proprietary labels from a private RAG corpus. It must remain usable without any private RAG folder.

## Public Release

When preparing a GitHub upload or public release, use only the public version and read the repo-level release/privacy check from the repository root at `references/release-and-privacy.md` first. Run a privacy check for private RAG folders, `.ftindex`, `.env`, credentials, personal paths, raw source files, screenshots, and private labels.

## Reference Reading Rule

When this skill selects a reference file, first scan the file structure, then read the sections and nearby guardrails relevant to the current task. Read the complete file when it is short, when the task is broad or strategic, or when partial reading could miss constraints. Do not rely on stale memory or heading-only scans. For publishing, still read the repo-level `references/release-and-privacy.md` carefully enough to complete the checklist.

## Workflow

1. Read the relevant areas of `references/experience.md` before deep analysis; read the full file when the task is broad or the active playbook may affect the answer.
2. During decomposition, actively expand stock, index, and theme questions into emotion-cycle, main-line/follower, expectation-gap, crowding, cross-market sentiment, or mainline/funds/game/cycle checks when these lenses can change the conclusion, even if the user did not explicitly request them. For collective sector surges, continuation questions, and market-sector-stock resonance, apply `Market-Sector-Stock Resonance And Continuation` from `references/sentiment-framework.md`; separate logic durability, tape continuity, and entry quality instead of treating a strong narrative as an automatic buy point. Read the relevant areas of `references/sentiment-framework.md`. Read the repo-level `references/release-and-privacy.md` before publishing or preparing a GitHub upload.
3. Gather or receive evidence from the market-specific skill first:
   - A-shares: prefer `cn-stock-move-reason` for quote, announcements, 股吧, board ranks, breadth, and A-share emotion cycle.
   - Japanese stocks: prefer `jp-stock-move-reason` for quote, news, Yahoo 掲示板, metrics, and theme/peer context.
   - US stocks/indexes: prefer `us-stock-move-reason` for why-up/why-down/mover questions, choose `us-stock-gamma-moomoo` for option/gamma questions, and choose `stock-technical-analysis` for chart/trend questions; use them together when catalyst, positioning, and price action all matter.
   - For U.S. community discussion evidence, `moomoo-comment-sentiment` can be used when installed. Treat it as a moomoo commu

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 6 — file references are metadata, not package contents.

View saved version →

---
name: stock-technical-analysis
description: Use when the user asks for technical analysis or a technically grounded entry-stop-exit plan for an individual US, Japanese, or A-share stock, including intraday trend, pressure/support, whether it can reach a price, K-line structure, volume-price behavior, moving averages, KDJ/MACD/RSI, Vegas channels, moomoo/Yahoo chart reads, US-only dark-pool level confirmation, or whether a breakout/pullback is confirmed.
---

# Stock Technical Analysis

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill when the user asks about price action, technical setup, intraday odds, support/resistance, trend continuation, pullback risk, or chart-based timing for US stocks, Japanese stocks, or A-shares.

This public-safe skill is self-contained and contains only generalized technical-analysis methods. Do not commit personal information, API keys, account data, private RAG files, or private research materials to GitHub. It must not store personal positions, private research paths, proprietary indicator names, private person names/handles, or private strategy labels. It may use a user-specified private RAG folder during a session, but the skill must remain usable without it.


## Public Release

When preparing a GitHub upload or public release, use the public version only and run the release/privacy check in this Skill at `references/release-and-privacy.md`. Never upload private RAG folders, `.ftindex` files, credentials, `.env`, personal data, screenshots, raw PDFs/PPTs, or private strategy labels.

## Core Rule

Before deep analysis, read `references/experience.md` if it exists, but only the `Active Playbook` and `Compression Protocol` sections unless the user explicitly asks for historical lessons. Apply those lessons to intraday execution realism, pressure/support confirmation, and post-discussion learning.

Cross-skill calls are operational. When this workflow says to use another market skill, actually load that skill's `SKILL.md` and required references if the skill is installed or available as a sibling in this repository. Do not merely mention the other skill by name in the answer.

When a live chart is visible in moomoo or another trading/chart app, use the chart directly if the user asks for `现在`, `再看看`, `moomoo`, `分时`, `图表`, `资金流`, `盘口`, or asks whether a level is being confirmed. Confirm the ticker and chart timeframe before reading signals.

For U.S. stocks and ETFs, prefer direct moomoo OpenD 1-minute K-line data over screenshot-only reads when OpenD is available. Subscribe to `SubType.K_1M` first, then call `get_cur_kline`; if the subscription fails because of missing permissions, unsupported symbols, or index restrictions, fall back to the moomoo app chart or a user-provided screenshot. OpenD may not support U.S. index K-lines such as `.SPX`, and some local setups may not have Japanese-stock or A-share permissions, so use SPY or the relevant ETF as a proxy only when appropriate and state the proxy clearly. When using SPY to judge SPX intraday structure, convert SPY levels into SPX levels before answering, using the freshest SPX or ES/SPX anchor available; report SPX levels first and note the SPY proxy only as source context.

For U.S. single-stock technical anomaly checks, use `moomoo-technical-anomaly` as an optional first-pass scanner when it is installed and OpenD is available. Treat its K-line, MA, MACD, RSI, KDJ, BOLL, and other abnormal-event output as a prompt for where to look, not as a buy/sell conclusion. The final technical judgment must still follow this skill's own framework: trend location, VWAP / moving-average relationship, volume-price confirmation, momentum divergence, support/resistance, failed breakout, and whether price holds after the signal.

Never answer only by following the latest tick. Use the sequence:

1.

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 6 — file references are metadata, not package contents.

View saved version →

---
name: us-stock-gamma-moomoo
description: Analyze US stock and ETF option gamma exposure with moomoo OpenD, plus .SPX/SPXW index-option structure using SPY/ES/CFD conversion when needed. Use for gamma dashboards or screenshots with authorized local references, gamma, GEX, gamma wall, gamma flip, SPX/SPY/ES intraday gamma, 0DTE option scenario value tables, option positioning, US-stock dark pool/off-exchange flow, borrow fee, FTD, short volume, or ChartExchange confirmation. Produces plain-language conclusions from supplied evidence or available option chains, snapshots, Greeks, OI, IV and prices; raw JSON is only for explicit export requests.
metadata:
  version: 0.1.14
---

# US Stock Gamma With moomoo

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill to turn moomoo OpenD option data into an actionable gamma map and a plain-language trading note.

This public-safe skill is self-contained. Do not commit personal information, API keys, account data, private RAG files, or any `Stocks` folder contents to GitHub. It may use a user-specified private RAG folder during a session, but it must not store private source paths, personal positions, original strategy names, private person names/handles, proprietary labels, or private document titles. Generic market concepts such as gamma, GEX, dealer hedging, FVG, KDJ, MACD, RSI, VWAP, and Vegas may be retained.


## Public Release

When preparing a GitHub upload or public release, use the public version only and run the release/privacy check in this Skill at `references/release-and-privacy.md`. Never upload `Stocks/`, private RAG folders, `.ftindex` files, credentials, `.env`, personal data, screenshots, raw PDFs/PPTs, or private strategy labels.

## Experience

Before deep analysis, read `references/experience.md` if it exists, but only `Active Playbook` and `Compression Protocol` unless the user explicitly asks for historical lessons. If a multi-turn correction produces a durable reusable lesson about gamma interpretation, option scenario handling, news/gamma interaction, or chart confirmation, apply it to the current answer. Update public experience only when the user explicitly requests a public Skill update; generalize the lesson and strip private details.

Cross-skill calls are operational. When this workflow says to use another market skill, actually load that skill's `SKILL.md` and required references if the skill is installed or available as a sibling in this repository. Do not merely mention the other skill by name in the answer.

Required coordination: for US option/gamma analysis, use this skill as the positioning and option-structure entry point, and add supporting skills based on the analysis workflow, not only on the user's wording. If the analysis is about SPX, SPXW, SPY, QQQ, Nasdaq, Dow, Russell, VIX, Nikkei proxy gamma, or any broad index/ETF gamma map, load `macro-news-check` because index gamma cannot be read well without current macro and broad-market tape. If the analysis discusses news acceptance, theme crowding, risk-on/risk-off tone, forum/social sentiment, or leader/follower context, load `stock-sentiment-analysis`. If the final view depends on support/resistance, intraday timing, trend confirmation, failed breakout, or price-action validation, load `stock-technical-analysis`. Gamma is the positioning map; macro, sentiment, and chart structure decide whether the map is being accepted or rejected.

When official moomoo anomaly skills are installed, use them as an auxiliary scan layer, not as a replacement for the self-calculated gamma map. For user requests involving `异动`, `大单`, `IV`, `PCR`, `聪明钱`, unusual options activity, option sentiment, or volatility anomaly, call `moomoo-derivatives-anomaly` alongside this skill by default. For U.S. stocks, do not request Hong Kong-only warrant / CBBC dimensions; us

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 15 — file references are metadata, not package contents.

View saved version →

---
name: us-stock-move-reason
description: Use when analyzing why a U.S. stock or ETF moved sharply, including premarket/after-hours gaps, earnings reactions, guidance, ratings, company news, unusual options, capital-flow anomalies, short data, community sentiment, and technical confirmation. Coordinates official moomoo skills and local market skills into a Chinese evidence-based move-reason note.
---

# US Stock Move Reason

先按[运行环境能力](references/runtime-capabilities.md)确认当前可用工具；下文工具与脚本流程在能力和授权具备时执行，不按宿主名称删减研究方法。

开始分析前，按[公共与私密 Reference 双读契约](references/reference-layers.md)读取本任务相关的公共与已获授权的私密 reference；两层均适用时都要读。

Use this skill as the U.S. stock counterpart to `jp-stock-move-reason` and `cn-stock-move-reason`. It is an evidence-gathering and synthesis workflow, not a trading bot.

This public-safe skill must not store account data, OpenD logs, API keys, cookies, private RAG paths, personal positions, screenshots, raw private notes, or proprietary labels. It may call official moomoo skills when installed, but those skills remain external data/anomaly providers.

## DTM API Context

Use the canonical JSON interfaces in `https://daytrading.monster/api-docs/` for DTM reads. Use `https://daytrading.monster/api/themes` without a market filter to trace cross-market industry-chain and theme transmission across US, Japan, and China. Read `themes[]` and `constituents[]` for members, `weight`, `reason_zh`, coverage, and completed-session returns; check dates instead of treating them as live moves. Parse the `text/plain` response bodies as JSON.

## Workflow

1. Normalize the target into a U.S. market symbol such as `US.NVDA`, `US.DELL`, `US.SPY`, or `US.TSLA`. If the user gives only a name and the listing is ambiguous, ask one concise question. Treat broad index questions (`SPX`, `SPY`, `QQQ`, `NQ`) as U.S. index/ETF workflows and consider `macro-news-check` by default.

2. Establish whether the move is real:
   - Use `moomooapi` when available for snapshot, premarket/after-hours fields, daily or 1-minute K-line context, volume, turnover, market state, and basic stock information.
   - If OpenD or permissions fail, use public quote sources only as fallback and state the limitation.
   - For SPX, moomoo may reject `US..SPX` index snapshots while still allowing SPX/SPXW option chains. Use `SPY`, ES/CFD, or user-provided SPX anchors when needed and state the anchor.

3. Gather confirmed catalysts first:
   - Use `moomoo-news-search` or `moomoo-stock-digest` for current company news, earnings, guidance, ratings, orders, regulatory events, M&A, capital actions, analyst notes, and sector read-through.
   - When earnings are involved, apply expectation gap explicitly: `prior market expectation` -> `actual result / guidance / commentary` -> `above expectation`, `in line or merely landed`, or `below expectation`.
   - Do not let social posts, option prints, or technical signals replace confirmed filings/news.

4. Scan official moomoo anomaly layers when they are installed and relevant:
   - `moomoo-capital-anomaly`: use for capital-flow, broker, short-sale, or funds-flow anomaly requests. A `无异常` response is a usable result.
   - `moomoo-derivatives-anomaly`: use for U.S.-applicable option dimensions only: `option_unusual`, `option_volatility`, `option_volume_price`, `option_sentiment`, and `option_comprehensive`. Do not request Hong Kong warrant / CBBC dimensions for U.S. stocks. If a full scan returns an opaque backend error, retry with explicit U.S. option dimensions before concluding the skill is unavailable.
   - `moomoo-technical-anomaly`: use as a first-pass technical anomaly scanner, then verify with `stock-technical-analysis`.
   - `moomoo-comment-sentiment`: use for moomoo community heat, disagreement, chasing, panic, and representative viewpoints. Label it as a moomoo community sample, not a full-market sentiment survey.

5. Add supporting skills based on what the evidence shows:
   - Use `us-stock-gamma-moomoo` when options positio

Preview of the first 4,000 characters. Read full instructions →

Referenced files: 4 — file references are metadata, not package contents.