Skip to content
analyze-jgb-insurer-superlong-flow logo

分析日本保險公司超長期 JGB 淨買賣流量 Skill

analyze-jgb-insurer-superlong-flow

從日本保險公司對超長期(10年以上)JGB 的淨買賣時間序列,自動產出「本月是否創紀錄淨賣出、連續淨賣出月數、期間累積淨賣出」等結論。

SKILL.md

Full skill instructions

分析日本保險公司超長期 JGB 淨買賣流量 Skill

以 JSDA 公開數據驗證「保險公司創紀錄賣超長端國債」等敘事,提供可複製的摘要(含 streak / record / 累積值)。

<essential_principles>

<principle name="jsda_data_source"> **核心數據來源:JSDA 公社債店頭売買高(Trading Volume of OTC Bonds)**

日本證券業協會(JSDA)自 2018/​05 起將投資人別交易統計整併進「Trading Volume of OTC Bonds」資料集。

數據位置:

關鍵 Sheet:

  • (J)合計差引 - 淨買賣額(Sell - Purchase)

數據特徵:

  • 頻率:月度
  • 分類:投資人類型 × 債券類型 × 天期桶
  • 單位:億日圓(100 million yen)
  • 延遲:約 T+1 個月

注意:2018/​05 前的舊版「Trends in Bond Transactions (by investor type)」已停止更新。 </​principle>

<principle name="sign_convention"> **符號慣例(重要!)**

JSDA 使用「賣出 - 買入」作為差引計算方式:

net_sale = 賣出金額 - 買入金額
  • 正值 = 淨賣出(賣出 > 買入,需求減少)
  • 負值 = 淨買入(買入 > 賣出,需求增加)

這與部分新聞報導的符號相反,需特別注意。 </​principle>

<principle name="maturity_bucket_mapping"> **天期桶口徑對齊**

JSDA Excel 欄位結構(第 4 列):

JSDA 欄位英文說明
超長期Interest-bearing Long-term (over 10-year)10 年以上利付債
利付長期Interest-bearing Long-term5-10 年利付債
利付中期Interest-bearing Medium-term2-5 年利付債
割引Zero-Coupon零息債
国庫短期証券等Treasury Discount Bills短期國庫券

本 Skill 使用:超長期(對應新聞常見的「10+ years」或「super-long」口徑) </​principle>

<principle name="investor_mapping"> **投資人分類**
JSDA 分類英文說明
生保・損保Life & Non-Life Insurance Companies壽險 + 產險合計
都市銀行City Banks大型商業銀行
地方銀行Regional Banks區域性銀行
信託銀行Trust Banks含年金管理
外国人Foreigners海外投資者

本 Skill 使用:生保・損保(保險公司合計) </​principle>

<principle name="record_detection"> **歷史極值判斷邏輯**
record_high = max(series)  # 最大淨賣出(正值最大)
is_record_sale = (latest == record_high) AND (latest > 0)

注意事項:

  • 數據起點會影響「歷史紀錄」的判定,輸出必須說明樣本期間
  • 若僅為近期極值,需標註「近 N 個月新高」 </​principle>

</​essential_principles>

<objective> 驗證「日本保險公司創紀錄賣超長端 JGB」等敘事,並提供: 1. **本月淨賣出/買入**:最新月份數值(億日圓) 2. **是否創紀錄**:判斷是否為樣本最大淨賣出(正值最大) 3. **連續淨賣出月數**:streak_len 4. **本輪累積淨賣出**:cum_over_streak 5. **Z-score 與分位數**:相對歷史的極端程度 </​objective>

<quick_start>

最快的方式:執行快速分析

cd .claude/​skills/​analyze-jgb-insurer-superlong-flow
pip install pandas numpy openpyxl  # 首次使用
python scripts/​jsda_flow_analyzer.py --quick

輸出範例(2025/​12 實測結果):

## 日本保險公司超長期 JGB 淨買賣驗證報告

**分析期間**:2022-04 ~ 2025-12(45 個月)

### 核心結論

| 指標 | 數值 | 說明 |
|------|------|------|
| 本月(2025-12)| **8,224 億日圓** | 淨賣出 |
| 是否創歷史紀錄 | **✓ 是** | 全樣本 (45 個月) |
| 連續淨賣出月數 | **5 個月** | 自 2025-08 起 |
| 本輪累積淨賣出 | **13,959 億日圓** | 1.40 兆日圓 |

### Headline Takeaways

1. ✓ 驗證屬實:日本保險公司在 2025/​12 創下歷史最大單月淨賣出
2. 已連續 5 個月淨賣出超長期國債,累積 1.40 兆日圓
3. 當前淨賣出規模處於歷史極端區間(Z-score: 2.71)

完整分析(含歷史比較):

python scripts/​jsda_flow_analyzer.py --full --format json

強制重新下載數據:

python scripts/​jsda_flow_analyzer.py --quick --refresh

</​quick_start>

<intake> 您想要執行什麼操作?
  1. 快速檢查 - 查看最新月份的淨買賣與連續淨賣出狀態
  2. 完整分析 - 執行完整的歷史比較與極值判斷
  3. 驗證新聞 - 輸入新聞的數字,對比 JSDA 原始數據
  4. JSON 輸出 - 輸出結構化 JSON 供後續處理

請選擇或直接執行分析。 </​intake>

<routing> | Response | Action | |---------------------------------|-------------------------------------| | 1, "快速", "quick", "check" | `python scripts/​jsda_flow_analyzer.py --quick` | | 2, "完整", "full", "analyze" | `python scripts/​jsda_flow_analyzer.py --full` | | 3, "驗證", "verify", "新聞" | 讀取 workflows/​verify-claim.md | | 4, "json", "JSON" | `python scripts/​jsda_flow_analyzer.py --format json` | </​routing>

<input_schema>

<parameter name="--quick" required="false"> **Type**: flag **Description**: 快速檢查模式(使用快取數據) </​parameter> <parameter name="--full" required="false"> **Type**: flag **Description**: 完整分析模式 </​parameter> <parameter name="--start-year" required="false" default="2018"> **Type**: int **Description**: 起始財年(日本財年 4 月開始) </​parameter> <parameter name="--lookback" required="false"> **Type**: int **Description**: 回溯月數(預設全樣本) </​parameter> <parameter name="--format" required="false" default="markdown"> **Type**: string **Options**: `json` | `markdown` **Description**: 輸出格式 </​parameter> <parameter name="--refresh" required="false"> **Type**: flag **Description**: 強制重新下載數據 </​parameter> <parameter name="--output" required="false"> **Type**: string **Description**: 輸出檔案路徑 </​parameter>

</​input_schema>

<output_schema>

JSON 輸出結構:

{
  "skill": "analyze_jgb_insurer_superlong_flow",
  "version": "1.0.0",
  "as_of": "2026-01-26",
  "data_source": {
    "name": "JSDA Trading Volume of OTC Bonds (公社債店頭売買高)",
    "url": "https://www.jsda.or.jp/​shiryoshitsu/​toukei/​tentoubaibai/",
    "sheet": "(J)合計差引"
  },
  "parameters": {
    "investor_group": "life_and_nonlife_insurance",
    "investor_label": "生保・損保 (Life & Non-Life Insurance Companies)",
    "maturity_bucket": "super_long",
    "maturity_label": "超長期 (Interest-bearing Long-term over 10-year)",
    "sign_convention": "正值=淨賣出 (Sell-Purchase), 負值=淨買入",
    "unit": "100 million yen (億円)"
  },
  "analysis_period": {
    "start": "2022-04",
    "end": "2025-12",
    "months": 45
  },
  "latest_month": {
    "date": "2025-12",
    "net_sale_100m_yen": 8224,
    "net_sale_trillion_yen": 0.8224,
    "interpretation": "淨賣出"
  },
  "record_analysis": {
    "is_record_sale": true,
    "record_sale_100m_yen": 8224,
    "record_sale_date": "2025-12",
    "record_purchase_100m_yen": -8889,
    "record_purchase_date": "2023-03",
    "lookback_period": "全樣本 (45 個月)"
  },
  "streak_analysis": {
    "consecutive_net_sale_months": 5,
    "streak_start": "2025-08",
    "cumulative_net_sale_100m_yen": 13959,
    "cumulative_net_sale_trillion_yen": 1.3959
  },
  "historical_stats": {
    "mean_100m_yen": -2025,
    "std_100m_yen": 3785,
    "latest_zscore": 2.71,
    "latest_percentile": 0.9778
  },
  "headline_takeaways": [
    "✓ 驗證屬實:日本保險公司在 2025/​12 創下歷史最大單月淨賣出(8,224 億日圓)",
    "已連續 5 個月淨賣出超長期國債,累積金額 13,959 億日圓(1.40 兆日圓)",
    "當前淨賣出規模處於歷史極端區間(Z-score: 2.71,超過 2 個標準差)"
  ]
}

</​output_schema>

<reference_index> 參考文件 (references/)

文件內容
data-sources.mdJSDA 數據來源與 XLS 下載說明
methodology.md計算方法論(streak、record、cumulative)
</​reference_index>

<scripts_index>

ScriptCommandPurpose
jsda_flow_analyzer.py--quick快速檢查
jsda_flow_analyzer.py--full完整分析
jsda_flow_analyzer.py--refresh強制重新下載數據
</​scripts_index>

<success_criteria> Skill 成功執行時:

  • 輸出最新月份淨賣出/買入數值
  • 判斷是否為歷史極值(含回溯期間說明)
  • 計算連續淨賣出月數與累積金額
  • 明確標示天期桶與投資人口徑
  • 提供 Z-score 與分位數
  • 可操作的 headline takeaways </​success_criteria>

<directory_structure>

analyze-jgb-insurer-superlong-flow/
├── SKILL.md                           # 本文件(主入口)
├── skill.yaml                         # 前端展示元數據
├── workflows/
│   ├── quick-check.md                 # 快速檢查工作流
│   ├── full-analysis.md               # 完整分析工作流
│   └── verify-claim.md                # 驗證新聞工作流
├── references/
│   ├── data-sources.md                # 資料來源說明
│   └── methodology.md                 # 方法論與公式
├── scripts/
│   └── jsda_flow_analyzer.py          # 主分析腳本(含數據下載)
└── data/
    └── cache/                         # 自動緩存目錄(.gitignore)

</​directory_structure>

More skills from majiayu000

xiaohongshu logo
majiayu000/claude-arsenal

xiaohongshu

xiaohongshu

286 148
View
agent-task-conductor logo
majiayu000/claude-skill-registry

agent-task-conductor

Conduct multi-agent task orchestration and workflow coordination.

663 1
View
conductor-setup logo
majiayu000/claude-skill-registry

conductor-setup

Initialize project with Conductor artifacts (product definition,

663 1
View
animation-designer logo
majiayu000/claude-skill-registry

animation-designer

Expert in web animations, transitions, and motion design using Framer Motion and CSS

663 1
View
diagramming logo
majiayu000/claude-skill-registry

diagramming

Creates Mermaid and ASCII diagrams for flowcharts, architecture, ERDs, state machines, mindmaps, and more. Use when user mentions diagram, flowchart, mermaid, ASCII diagram, text diagram, terminal diagram, visualize, C4, mindmap, architecture diagram, sequence diagram, ERD, or needs visual docume...

663 1
View
h3-pg logo
majiayu000/claude-skill-registry-data

h3-pg

PostgreSQL bindings for H3 hexagonal grid system. Use when working with H3 cells in Postgres, including spatial indexing, geometry/geography integration, and raster analysis.

23 1
View
conductor-development logo
majiayu000/claude-skill-registry

conductor-development

Context-Driven Development skill for projects using Conductor. Use this skill when you detect a `conductor/` directory in the project, when working on tasks defined in a `plan.md` file, or when the user asks about tracks, specs, or plans. Automatically applies TDD workflow, tracks task completion...

663 1
View
conductor-status logo
majiayu000/claude-skill-registry

conductor-status

Display project status, active tracks, and next actions

663 1
View
dockerization logo
majiayu000/claude-skill-registry

dockerization

Official Stakpak application containerization standard operating procedure, a step-by-step guidline to properly dockerize applications. This is a rule book curated by the Stakpak Team.

663 1
View

Popular AI tools

Kaiber logo
Video

Kaiber

Generate, edit, and beat-sync AI video with leading models in one workspace.

Paid
View
Vimcal logo
Productivity

Vimcal

The world's fastest calendar for remote work

Free
View

Transform Your Design with AI Designer by ImgCreator.ai

Freemium
View
Akool AI logo
Content & writing

Akool AI

Revolutionizing Video Production with AI-Powered Creativity

Paid
View

Extend an image past the frame and let AI fill the new aspect ratio.

Freemium
View
StarByFace logo
Security

StarByFace

Discover your celebrity doppelgänger with StarByFace!

Free
View
C

ChainClarity explains 700+ crypto whitepapers in plain English, with layered summaries, comparisons, research tools, alerts, and a $4.99 Pro plan.

Freemium
View
Opus Clip logo
Coding & apps

Opus Clip

Opus.ai: Revolutionize Your Web Experience

Free
View