zeberg-salomon-rotator
用領先和同時指標建構景氣模型,並判讀兩種景氣階段:景氣擴張期 (Risk-On) 與景氣收縮期 (Risk-Off),並根據理論給出 "撞冰山" 與 "下沉" 事件訊號。
Profile →npx skills add majiayu000/claude-skill-registry --skill zeberg-salomon-rotator --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
<essential_principles> <principle name="two_state_model"> **兩態切換模型核心** Zeberg–Salomon 模型將市場簡化為兩種狀態: - **RISK_ON**: 持有股票(SPY),景氣擴張期 - **RISK_OFF**: 持有長債(TLT),景氣收縮期 切換邏輯基於「領先指標先轉弱,同時指標後確認」的景氣循環規律。 </principle> <principle name="leading_coincident"> **領先 vs 同時指標** | 類型 | 作用 | 典型成分 | 領先時間 | |------------|------|------------------------------|----------| | Leading | 預警 | 殖利率曲線、新訂單、房市許可 | 6-12 月 | | Coincident | 確認 | 就業、工業生產、實質收入 | 同步 | **合成方式**: 1. 各序列做 transform(yoy/mom/diff) 2. 統一方向(direction +1/-1) 3. Rolling z-score 標準化 4. EMA 平滑 5. 加權合成 </principle> <principle name="iceberg_sinking"> **冰山事件 vs 下沉事件** ``` Iceberg Event: LeadingIndex < iceberg_threshold → 預警:景氣開始轉弱 → 搭配「領先指標下降」+ 可選「市場亢奮」濾鏡 Sinking Event: CoincidentIndex < sinking_threshold → 確認:實體經濟收縮 → 通常在 Iceberg 之後數月發生 ``` **狀態機邏輯**: - RISK_ON → RISK_OFF:Iceberg 連續確認 + 斜率為負 - RISK_OFF → RISK_ON:領先指標回升超過 (threshold + hysteresis) </principle> <principle name="data_access"> **資料取得方式** 本 skill 使用**無需 API key** 的資料來源: - **FRED CSV**: `https://fred.stlouisfed.org/graph/fredgraph.csv?id={SERIES_ID}` - **Yahoo Finance**: `yfinance` 套件抓取 SPY, TLT, VIX 腳本位於 `scripts/` 目錄,可直接執行。 </principle> </essential_principles> <objective> 實作 Zeberg–Salom
cd skills/zeberg-salomon-rotator pip install pandas numpy yfinance pandas-datareader # 首次使用 python scripts/rotator.py --quick python scripts/rotator.py --start 2000-01-01 --end 2026-01-01 --output result.json
What does the zeberg-salomon-rotator skill do?
用領先和同時指標建構景氣模型,並判讀兩種景氣階段:景氣擴張期 (Risk-On) 與景氣收縮期 (Risk-Off),並根據理論給出 "撞冰山" 與 "下沉" 事件訊號。
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill zeberg-salomon-rotator --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.