← Back to Skills Marketplace
cjlrestlong-ai

Hk Stock Morning Report

by cjlrestlong-AI · GitHub ↗ · v1.2.7 · MIT-0
cross-platform ⚠ suspicious
347
Downloads
0
Stars
0
Active Installs
25
Versions
Install in OpenClaw
/install hk-stock-morning-report
Description
Generate HK stock market morning report (股市晨報) for Chinese bank trading desk. Use when user asks "生成晨报", "股市晨报", "今日股市", "港股晨報", or any similar HK stock mark...
README (SKILL.md)

HK Stock Morning Report (股市晨報)

⚠️ 第四部分:熱門港股

搜索關鍵詞格式(強制): {報告日前一日}熱門股票{報告日前一日}港股熱點

  • 動態替換日期,例如報告日4月28日 → 搜索「4月27日熱門股票」
  • 不是搜特定股票,是搜「熱門股票」這個主題,找到有新聞的股票後選1檔
  • 自己根據搜索結果編寫新聞摘要,不直接引用未經核查的數字
  • 搜不到具體新聞時:誠實寫「待更新」,不得編造數字

🚨 強制規則(零藉口)

  1. 第一部分不准猜測假日 — 必須先確認當日是否交易日,不準主觀以為是假期
    • 香港2026年假日查證:只查 https://www.gov.hk/sc/about/abouthk/holiday/2026.htm
    • 佛誕翌日是5月26日(不是4月),耶穌受難節是4月3日(不是4月27日)
  2. 第三部分搜索必須加「年月日」 — 關鍵詞格式:YYYY年MM月DD日 港股/財經新聞
    • 例:4月28日晨報 → 搜索「2026年4月28日 港股 新聞」
    • 不加日期會返回舊聞,導致數據時間錯乱
  3. 早上8點,只可用上一個交易日的收盤數據 — 港股9:30開盤,8點不可能有今天收盤數;上一個交易日(4/27)收盤數據可用,今天(4/28)收盤數據不可用
  4. 數據內部溯源(不上螢幕報告) — 每個引用數字必須記錄來源URL和時間戳,存在工作記憶中備查;報告本身保持乾淨,不展示時間戳
  5. 發送前二次核查清單 — 推送前必須逐項確認:
    • 今天是什麼日期?→ 大標題
    • 上一個交易日是哪天?→ 第一部分標題
    • 今天是否交易日?→ 查gov.hk
    • 搜索關鍵詞含「YYYY年MM月DD日」?
    • 早上(≤9:30)不得有今天收盤數據?
    • 每個數字有靠譜來源?懷疑的數字有二次核查?
    • 南下資金口徑確認(成交?凈買入?)
    • 標題和摘要之間有空行?
    • 報告已讀一遍,確認數字邏輯正確?

固化文件(直接引用,不重新揣測)

文件 用途
scripts/fetch_southbound_data.py 南下資金數據自動獲取腳本,輸出JSON
scripts/southbound_report_template.md 完整模板+格式規範+核查清單
references/errors.md 錯誤全集,發送前必查
references/stock_report_format.md 格式模板原文

南下資金自動化流程(v1.0.6+)

  1. 搜索:用 tavily_search 搜索 site:stcn.com 南向資金 {日期} 凈買入
  2. 確認口徑:stcn.com 數據寶頁面會同時有「成交額」和「凈買入」,兩者不要混淆
    • 全市場凈買入 = 「合計凈買入」那個數字(不是成交活躍股數字)
  3. 寫入 JSON:把搜索結果(口徑+日期+數字)寫入 data/southbound_latest.json
  4. 格式化:運行 python3 scripts/fetch_southbound_data.py 讀取 JSON 並輸出格式化文字
  5. 核查:確認數字口徑是全市場,不是成交活躍股

關鍵區分(stcn.com 頁面):

  • stcn.com 同一個頁面同時有:
    • 「成交活躍股合計成交222.26億,凈買入22.51億」→ 這是個股口徑
    • 「合計成交額826.54億,凈買入37.38億」→ 這是全市場口徑
  • 晨報必須用全市場口徑(37.38億),不是成交活躍股口徑(22.51億)

⚠️ 2026-04-22 錯誤檢討後更新的核心規則(v1.0.5+)

南下資金數據口徑規則(最高優先)

stcn.com同時有兩套口徑,必須區分:

  • 「港股通成交活躍股」口徑 → 這是個股凈買入,不是全市場總數
  • 「南向資金今日凈買入XX億」口徑 → 這才是全市場總數,晨報必須用這個

當看到「成交活躍股」字樣,立刻警惕:那不是我們需要的數字。

南下資金標準格式:

📍二、南下資金凈買入XX億(港元)
📈 港股通(滬): 凈買入YY億(港元)
📉 港股通(深): 凈賣出ZZ億(港元)

其中:凈買入XXX;

凈賣出XXX。

注意:無「成交」字段,只顯示凈買入/凈賣出金額。口徑必須是全市場「合計凈買入」數字,不是成交活躍股數字。

用數字前三驗原則

每次使用任何數字前,必須確認三項:否則寫「待更新」:

  1. 口徑:這是成交額?凈買入?不同口徑數字差異巨大
  2. 日期:這個數字是哪天的?不是目標日期的立刻排除
  3. 來源:這個數字來自哪個網站?不同來源口徑不同

歷史版本優先原則

  • 當用戶說「某版本有完整數據」,先查歷史消息確認內容,再決定是直接使用還是修補
  • 不得用新的搜索結果直接覆蓋可能正確的歷史版本

勝宏科技代碼

  • 02476.HK(2026-04-22 更新,之前錯為03839)
  • 搜索確認,不得猜測

Workflow (执行步骤)

Step 1: Read Format Template

Read references/stock_report_format.md — this is the source of truth for all format rules. Do not deviate.

Step 2: Determine Date Header

⚠️ Big title date (x.xx) = Report generation date (natural day), NOT last trading day

Example: Report generated on Sunday Apr 19 but covering Friday Apr 18 trading → big title = 4.19, Section 1 = 上週五股市回顧

Find the last trading day and determine the Section 1 header:

  • Same week, no gap → 📍一、昨日股市回顧
  • Same week, gap (holiday) → 📍週X股市回顧
  • Last week → 📍一、上週X股市回顧

Step 3: Fetch Index Data

  • Hang Seng Index: https://qt.gtimg.cn/q=r_hkHSI
    • Field[3] = current price, Field[4] = previous close
    • Calculate: change points = price - prev_close; change% = (price-prev_close)/prev_close × 100
  • Hang Seng Tech Index: https://qt.gtimg.cn/q=r_hkHSTECH (same parsing)

Step 4: Search Section 1 (Market Review)

  • Search: site:gelonghui.com "港股收評" "昨日日期"
  • Extract: market overview (30-50 chars), strong/weak stocks with exact figures, sector performance
  • Fallback: futunn or Yahoo Finance

Step 5: Search Section 2 (Southbound Capital) — ⚠️ 最重要步驟

只使用stcn.com全市場口徑「南向資金今日凈買入XX億」

  • Primary: site:stcn.com "南向資金" "昨日日期" "凈買入"
  • Also: site:gelonghui.com "南向資金" "昨日日期"
  • Fallback: site:futunn.com "南下" "昨日日期"
  • ⚠️ stcn.com的「成交活躍股」數字不是全市場總數,絕對不能使用
  • ⚠️ 當細分(滬/深)數字不確定時:保留📈📉兩行並填入「待更新」,不得刪除行

Step 6: Search Section 3 (Hot News) — ⚠️ Search TODAY's date

  • Search today's date + "港股" + "熱點" or "最新"
  • Must search today's news, NOT last trading day's market data
  • Even if HK market is closed, there are macro policy, international market, broker reports with today's date
  • If no results → write "待更新", do NOT fabricate data
  • Select 3 most relevant news items

Step 7: Search Section 4 (Hot HK Stock News) — Select 1 stock, write your own summary

⚠️ 第四部分是新聞,新聞等於搜「熱門股票」而不是搜個股!

  • Search: "{報告日前一日}熱門股票""{報告日前一日}港股熱點"(例如:報告日4月23日 → 搜索「4月22日熱門股票」)
  • 不是搜特定股票名稱或代碼 — 搜的是「熱門股票」這個主題
  • 找出有最新新聞的股票,選1檔
  • 自己根據搜索結果編寫新聞摘要,不是直接引用未經核查的數字
  • 如果搜不到具體新聞 → 如實寫「待更新」,不得編造數字
  • 勝宏科技代碼:02476.HK(僅供參考,不強制使用)

Step 8: Generate Report

Assemble the report strictly following the format template. Big title = 3 Section 3 ▶️ titles joined by semicolons. Write 待更新 if data unavailable — never estimate.

Step 9: Pre-Send Verification

Read references/errors.md and verify each item. Check:

  • Big title = Section 3 ▶️ titles (do not rewrite)
  • Arrow direction: 📈=net buy, 📉=net sell
  • No estimated numbers (write 待更新 if uncertain)
  • Format: 📍 has no space after, ▶️ has no bold, Section 4 has only 1 stock, ▶️ title and content must have an empty line between them
  • 📈📉 rows with uncertain data: keep the rows, write "待更新", do NOT delete the rows
  • 南下資金口徑確認:數字來自全市場口徑,不是成交活躍股口徑
  • 勝宏科技代碼:02476.HK(搜索確認)

Step 10: Send

飛書只發卡片即可(2026-04-24 更新):

  • 卡片:完整晨報內容,header為「🔴股市晨報」(飛書卡片現已支援文字選取複製,無需再發代碼塊)

微信個人:純文字格式(不需卡片)

香港政府假日官網

  • 原則:查香港假日資訊,只使用香港政府官網
  • 官網:https://www.gov.hk/sc/about/abouthk/holiday/2026.htm
  • 示例:復活節日期(4月5日,不是4月22日)、核證交易日等,均以官網為準

Key Format Rules (7 Rigid Constraints)

  1. Big title contains 🔴 and 🔵
  2. 📍一: X = 昨日 / 週一週五 / 上週一上週五
  3. 📍二: Direction from actual data (📈=net buy, 📉=net sell)
  4. Numbers with(港元): only the first line of Section 2 header (e.g., 凈買入xx億(港元)), the 📈📉 lines and summary do NOT repeat (港元)
  5. ▶️ Title and content must have an empty line between (Error 9)
    • ⚠️ 全域段落空行規則(v1.2.3+):不只▶️與摘要之間,所有邏輯段落之間都必須有空行。這包括:大標題↔免責聲明、免責聲明↔📍一、📍N↔正文 ▶️標題↔摘要、資料來源↔版權聲明。微信/飛書的 Markdown 只識別 \ \ (雙換行)為空行,單換行無效。
  6. Section 4: only 1 stock
  7. 📍 followed directly by text, no space

Data Sources

Content Source
Index (close/change) Tencent Finance API qt.gtimg.cn
Section 1 (market overview) gelonghui.com 港股收評
Section 2 (southbound) stcn.com 全市場口徑(primary);gelonghui.com;futunn.com (fallback)
Section 3 (hot news) Web search for today's date
Section 4 (top stock) Web search for today's individual stock news
Usage Guidance
Do not run this skill in production or give it broad automation rights yet. Specific recommendations: - Missing files: SKILL.md references scripts/fetch_southbound_data.py and a southbound template that are not in the package — ask the author for the missing files or a corrected package. - Messaging: The skill instructs sending via WeChat and Feishu but declares no credentials or APIs — clarify how messages are delivered and where credentials should be stored. Never paste tokens into chat; prefer platform-integrated connectors. - External tool: SKILL.md requires 'tavily_search' (and web scraping of stcn.com etc.) but the skill does not declare that dependency. Confirm the runtime environment will provide the search tool or provide an alternate, auditable implementation. - Code issues: scripts/generate_report.py contains bugs (uses datetime.timedelta but imports datetime incorrectly) and assumes WORKDIR '/root/.openclaw/workspace'. Run the script in a sandbox and review/patch it before use; change file paths to a non-root workspace. - Data handling: the skill tells the agent to record source URLs/timestamps in memory (not shown how). Ensure any stored data does not leak to external services and that source collection follows your org's data policies. If you still want to use it: request the missing helper scripts and documentation for message delivery, test locally in an isolated environment, and require explicit configuration for any credentials with least privilege. If the author cannot explain the missing pieces and delivery method, treat the package as incomplete and untrusted.
Capability Analysis
Type: OpenClaw Skill Name: hk-stock-morning-report Version: 1.2.7 The skill bundle is a specialized utility for generating Hong Kong stock market reports for a financial trading desk. The Python script (generate_report.py) and instructions (SKILL.md) focus on fetching legitimate market data from the Tencent Finance API and specific financial news sites. There is no evidence of data exfiltration, malicious code execution, or harmful prompt injection; the strict formatting rules and 'mandatory' instructions are clearly designed to ensure data accuracy and professional output quality.
Capability Assessment
Purpose & Capability
The skill's stated purpose (generate HK morning report) matches the included format, rules, and a small generator script, but the SKILL.md also depends on other helper scripts (e.g., scripts/fetch_southbound_data.py, southbound_report_template.md) and an external search tool (tavily_search) that are not present in the package or declared as requirements. The skill also instructs sending via WeChat and Feishu yet declares no credentials or integration code.
Instruction Scope
Runtime instructions require: web searches (tavily_search), scraping specific sites (stcn.com, gelonghui, futunn, gov.hk), recording source URLs/timestamps in memory, and sending reports to WeChat personal + Feishu group. The SKILL.md gives the agent broad discretion to perform searches and collect data, but does not declare the search tool or any messaging integration; it also references writing JSON to data/southbound_latest.json and other files that are not present. This grants the agent open-ended network and filesystem actions that are not fully implemented in the package.
Install Mechanism
No install spec (instruction-only plus one script) — lowest install risk. The only included code is a local Python script that fetches index data from qt.gtimg.cn; there is no download-from-URL or package install. However, the script assumes certain runtime environment (python3 available, network access).
Credentials
The skill declares no required environment variables or credentials, yet SKILL.md expects the agent to send messages to WeChat and Feishu and to run tavily_search. Those delivery/send steps normally require API credentials/tokens. The absence of declared credentials is an inconsistency: either the skill expects the agent to have other integrations preconfigured (not documented) or it omits required secret access, which could lead to manual workaround behaviors or accidental exposure when the agent attempts to use other credentials.
Persistence & Privilege
always=false (normal). The included script writes output to WORKDIR = '/root/.openclaw/workspace' which suggests assumptions about filesystem layout and privileged user context; that path is unusual for a non-privileged skill and could overwrite files in a root home if run as root. The skill does not request persistent automatic inclusion or modify other skills' configs.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install hk-stock-morning-report
  3. After installation, invoke the skill by name or use /hk-stock-morning-report
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.7
v1.2.7 新增規則5:發送前二次核查清單——必須逐項確認8個條件後才能推送
v1.2.6
v1.2.6 新增規則4:數據內部溯源——每個引用數字必須記錄來源URL和時間戳,工作記憶中備查;報告本身保持乾淨,不展示時間戳
v1.2.5
v1.2.5 修正第三條規則:並非禁止收盤數據,而是早上8點只可用『上一個交易日』收盤數據,不可有『今天』的收盤數據(今天還沒收盤)。
v1.2.4
v1.2.4 三大致命錯誤:①自作主張以為4/27是佛誕假期(2026佛誕是5/25);②引用Yahoo晚間收盤數據充作早上可用數據;③第三部分搜索未加『年月日』導致返回舊聞。已新增強制規則:①第一部分不准猜測假日;②第三部分搜索必須含『年月日』;③早上8點不得有收盤數據。
v1.2.3
v1.2.3: 新增全域段落空行規則(錯誤9變體),所有邏輯塊之間必須使用雙換行分隔
v1.2.2
修正方向錯誤:港字應移除而非補回,統一為「昨日/週X/上週X股市回顧」(無港字)
v1.2.1
全面修補所有文件中'昨日股回顧'→'昨日港股回顧'的'港'字缺失錯誤(stock_report.js、generate_report.py、SKILL.md、job_steps等)
v1.2.0
固化方法論:確認正確晨報輸出流程(第一部分搜索原則、第三部分實時搜索原則)
v1.1.6
v1.1.6: 第四部分搜索詞明確為「報告日前一日」熱門股票,非固定日期
v1.1.5
v1.1.5: 第四部分改為搜索「熱門股票」主題+自行編寫新聞摘要,不再搜索特定個股
v1.1.4
v1.1.4: 移除SKILL.md南下資金格式中的「成交」字段,只保留「凈買入XX億(港元)」
v1.1.3
# hk-stock-morning-report 版本更新日誌 > 記錄每一次迭代變化。訪問者可在 [clawhub.com](https://clawhub.ai/skills/hk-stock-morning-report) 查看完整版本歷史。 --- ## 📋 版本更新日誌 ### 🆕 v1.1.2 — 2026-04-23 **作風調整**:日後遇到不明確環節,主動先向用戶確認再做,不再自行猜測 ### 🆕 v1.1.1 — 2026-04-23 **新增 CHANGELOG.md 版本歷史文件**,記錄 v1.0.0 至 v1.1.0 完整迭代 - 包含:每次版本更新內容、核心教訓、版本對照表 - 訪問者可在 clawhub.com 發布頁查看完整迭代歷程 ### 🐛 v1.0.8 — 2026-04-23 **修正格式錯誤** ❗重要更新 - 問題:輸出包含了「成交 1256.56億」,但用戶只要求「凈買入/賣出」金額 - 修正:移除「成交」字段,只保留「凈買入XXX億(港元)」 - 同步更新:fetch_southbound_data.py 腳本 + SKILL.md ### ⚙️ v1.0.7 — 2026-04-23 **固化 agent 三步流程** ❗重要更新 - 流程:tavily 搜索 → 確認口徑 → 寫入 JSON → 運行格式化腳本 - 修復:dict 比對 bug(計算 net_buy minus net_sell) - 同步:stcn.com 全市場口徑 vs 成交活躍股口徑區分寫入 SKILL.md ### 🔧 v1.0.6 — 2026-04-23 **從規則升級到工具** - 新增:fetch_southbound_data.py(格式化輸出腳本) - 新增:southbound_report_template.md(完整模板 + 核查清單) - 意義:從「文字規則」升級為「自動化工具」,減少主觀判斷 ### 📝 v1.0.5 — 2026-04-22 **錯誤教訓固化** ❗重要更新 - 新增:南下資金口徑規則(stcn.com 全市場 ≠ 成交活躍股,兩者差 4 倍) - 新增:三驗原則(口徑 + 日期 + 來源,每次用數字前必須確認) - 新增:歷史版本優先原則(用戶說某版本有數據,先查再使用) - 修正:勝宏科技代碼 02476.HK(之前錯為 03839.HK) ### 🔵 v1.0.0-v1.0.4 **初期框架建設** - v1.0.0:4 部分晨報結構(市場回顧 → 南下資金 → 熱點資訊 → 熱門港股) - v1.0.1:Tushare 停用,熱點搜索改為當天日期 - v1.0.2:新增虛擬資產 ETF 每日 9:00 監控任務(13 隻 ETF 清單) - v1.0.3:格式固化,飛書卡片 + 代碼塊雙版本推送 - v1.0.4:錯誤檢討口號階段(香港假日 gov.hk 規則) --- ## 📊 核心教訓 | 教訓 | 發生版本 | 影響 | |------|---------|------| | 數據口徑必須區分 | v1.0.5 | 全市場213.60億 vs 成交活躍股48.90億,差4倍 | | 成交金額 ≠ 凈買入 | v1.0.8 | 成交1256.56億 vs 凈買入213.60億 | | 勝宏科技代碼不可猜測 | v1.0.5 | 02476.HK(對)vs 03839.HK(錯)| | 大標題日期 ≠ 交易日期 | v1.0.5 | 大標題=報告生成日;第一部分=上一交易日 | | ▶️標題和摘要之間必須空一行 | v1.0.4 | 連續3次錯誤(錯誤9,最高頻)| | 飛書必須同時發卡片+代碼塊 | v1.0.4 | 之前只發了卡片漏了代碼塊 |
v1.1.2
## 📋 版本更新日誌 ### 🆕 v1.1.1 — 2026-04-23 **作風調整**:日後遇到不明確環節,主動先向用戶確認再做,不再自行猜測 ### 🆕 v1.1.0 — 2026-04-23 **新增 CHANGELOG.md 版本歷史文件**,記錄 v1.0.0 至 v1.1.0 完整迭代 - 包含:每次版本更新內容、核心教訓、版本對照表 - 訪問者可在 clawhub.com 發布頁查看完整迭代歷程 ### 🐛 v1.0.8 — 2026-04-23 **修正格式錯誤** ❗重要更新 - 問題:輸出包含了「成交 1256.56億」,但用戶只要求「凈買入/賣出」金額 - 修正:移除「成交」字段,只保留「凈買入XXX億(港元)」 - 同步更新:fetch_southbound_data.py 腳本 + SKILL.md ### ⚙️ v1.0.7 — 2026-04-23 **固化 agent 三步流程** ❗重要更新 - 流程:tavily 搜索 → 確認口徑 → 寫入 JSON → 運行格式化腳本 - 修復:dict 比對 bug(計算 net_buy minus net_sell) - 同步:stcn.com 全市場口徑 vs 成交活躍股口徑區分寫入 SKILL.md ### 🔧 v1.0.6 — 2026-04-23 **從規則升級到工具** - 新增:fetch_southbound_data.py(格式化輸出腳本) - 新增:southbound_report_template.md(完整模板 + 核查清單) - 意義:從「文字規則」升級為「自動化工具」,減少主觀判斷 ### 📝 v1.0.5 — 2026-04-22 **錯誤教訓固化** ❗重要更新 - 新增:南下資金口徑規則(stcn.com 全市場 ≠ 成交活躍股,兩者差 4 倍) - 新增:三驗原則(口徑 + 日期 + 來源,每次用數字前必須確認) - 新增:歷史版本優先原則(用戶說某版本有數據,先查再使用) - 修正:勝宏科技代碼 02476.HK(之前錯為 03839.HK) ### 🔵 v1.0.0-v1.0.4 **初期框架建設** - v1.0.0:4 部分晨報結構(市場回顧 → 南下資金 → 熱點資訊 → 熱門港股) - v1.0.1:Tushare 停用,熱點搜索改為當天日期 - v1.0.2:新增虛擬資產 ETF 每日 9:00 監控任務(13 隻 ETF 清單) - v1.0.3:格式固化,飛書卡片 + 代碼塊雙版本推送 - v1.0.4:錯誤檢討口號階段(香港假日 gov.hk 規則)
v1.1.1
## 📋 版本更新日誌 ### 🆕 v1.1.1 — 2026-04-23 **作風調整**:日後遇到不明確環節,主動先向用戶確認再做,不再自行猜測 ### 🆕 v1.1.0 — 2026-04-23 **新增 CHANGELOG.md 版本歷史文件**,記錄 v1.0.0 至 v1.1.0 完整迭代 - 包含:每次版本更新內容、核心教訓、版本對照表 - 訪問者可在 clawhub.com 發布頁查看完整迭代歷程 ### 🐛 v1.0.8 — 2026-04-23 **修正格式錯誤** ❗重要更新 - 問題:輸出包含了「成交 1256.56億」,但用戶只要求「凈買入/賣出」金額 - 修正:移除「成交」字段,只保留「凈買入XXX億(港元)」 - 同步更新:fetch_southbound_data.py 腳本 + SKILL.md ### ⚙️ v1.0.7 — 2026-04-23 **固化 agent 三步流程** ❗重要更新 - 流程:tavily 搜索 → 確認口徑 → 寫入 JSON → 運行格式化腳本 - 修復:dict 比對 bug( → 計算) - 同步:stcn.com 全市場口徑 vs 成交活躍股口徑區分寫入 SKILL.md ### 🔧 v1.0.6 — 2026-04-23 **從規則升級到工具** - 新增:fetch_southbound_data.py(格式化輸出腳本) - 新增:southbound_report_template.md(完整模板 + 核查清單) - 意義:從「文字規則」升級為「自動化工具」,減少主觀判斷 ### 📝 v1.0.5 — 2026-04-22 **錯誤教訓固化** ❗重要更新 - 新增:南下資金口徑規則(stcn.com 全市場 ≠ 成交活躍股,兩者差 4 倍) - 新增:三驗原則(口徑 + 日期 + 來源,每次用數字前必須確認) - 新增:歷史版本優先原則(用戶說某版本有數據,先查再使用) - 修正:勝宏科技代碼 02476.HK(之前錯為 03839.HK) ### 🔵 v1.0.0-v1.0.4 **初期框架建設** - v1.0.0:4 部分晨報結構(市場回顧 → 南下資金 → 熱點資訊 → 熱門港股) - v1.0.1:Tushare 停用,熱點搜索改為當天日期 - v1.0.2:新增虛擬資產 ETF 每日 9:00 監控任務(13 隻 ETF 清單) - v1.0.3:格式固化,飛書卡片 + 代碼塊雙版本推送 - v1.0.4:錯誤檢討口號階段(香港假日 gov.hk 規則)
v1.1.0
v1.1.0 | 2026-04-23 更新 CHANGELOG.md 版本歷史,記錄 v1.0.0-v1.0.9 完整迭代。 v1.0.9 | 2026-04-23 新增 CHANGELOG.md,包含版本迭代、核心教訓與對照表。 v1.0.8 | 2026-04-23 修正格式:移除「成交」字段,只保留「凈買入/賣出」金額。 v1.0.7 | 2026-04-23 agent搜索→JSON→format三步固化,修復dict bug。 v1.0.6 | 2026-04-23 新增 fetch_southbound_data.py + southbound_report_template.md。 v1.0.5 | 2026-04-22 口徑規則、三驗原則、歷史版本優先、代碼02476固化。 v1.0.0-v1.0.4 | 初期框架、格式規範、ETF監控任務。
v1.0.9
Version 1.0.9 - Added a CHANGELOG.md file to the project for documenting changes. - No changes made to skill logic or workflow.
v1.0.8
Version 1.0.8 - No file changes detected in this release. - All existing core workflows and format rules remain as previously specified. - No update to functionality, file structure, or documentation content.
v1.0.7
- Updated workflow for southbound capital data: now uses tavily_search to query stcn.com and confirms use of full-market net buy numbers only. - Clarified the distinction between "individual stock" vs "full market" data on stcn.com, mandating exclusive use of full market figures in the morning report. - Outlined the process of saving search results into JSON and formatting them via the provided script. - Added specific stepwise instructions to prevent use of "active stock" numbers and ensure consistency of reporting standards. - No code changes; documentation and procedure improvements only.
v1.0.6
- Added automated workflow for Southbound capital data using fetch_southbound_data.py, with direct use of JSON output and fallback to mock data marked as "待更新". - Clarified references to all format, template, and error check files—always use canonical documents, no re-interpreting. - Upgraded Southbound data process as v1.0.6: prioritize code/data automation; mock/fallback rules strictly stated. - All original reporting steps, data origin rules, and verification workflow remain, with clearer separation between format rules and implementation details.
v1.0.5
Version 1.0.5 Changelog - Strengthened rules for southbound capital flow section: **clearly distinguish stcn.com全市場口徑 from 個股成交活躍股數字; only use the former for section 2.** - Added explicit "三驗原則" before using any data (check type, date, and source) and require "待更新" if any are unclear. - Enforced code update for 勝宏科技: must use 02476.HK, not previous/guessed ticker. - Updated pre-send checklist and workflow steps with enhanced data validation and historic version reuse. - Clarified and highlighted key format and data source requirements throughout document.
Metadata
Slug hk-stock-morning-report
Version 1.2.7
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 25
Frequently Asked Questions

What is Hk Stock Morning Report?

Generate HK stock market morning report (股市晨報) for Chinese bank trading desk. Use when user asks "生成晨报", "股市晨报", "今日股市", "港股晨報", or any similar HK stock mark... It is an AI Agent Skill for Claude Code / OpenClaw, with 347 downloads so far.

How do I install Hk Stock Morning Report?

Run "/install hk-stock-morning-report" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Hk Stock Morning Report free?

Yes, Hk Stock Morning Report is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Hk Stock Morning Report support?

Hk Stock Morning Report is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Hk Stock Morning Report?

It is built and maintained by cjlrestlong-AI (@cjlrestlong-ai); the current version is v1.2.7.

💬 Comments