quantwise 1.2.0 → 1.2.2

package/.claude/skills/uptrend-analyzer/scripts/uptrend_analyzer.py

@@ -0,0 +1,219 @@
+#!/usr/bin/env python3
+"""
+Uptrend Analyzer - Main Orchestrator
+
+Analyzes market breadth using Monty's Uptrend Ratio Dashboard data.
+Generates a 0-100 composite score from 5 components:
+1. Market Breadth (Overall) - 30%
+2. Sector Participation - 25%
+3. Sector Rotation - 15%
+4. Momentum - 20%
+5. Historical Context - 10%
+
+No API key required - uses free GitHub CSV data.
+
+Usage:
+    python3 uptrend_analyzer.py
+    python3 uptrend_analyzer.py --output-dir /path/to/output
+
+Output:
+    - JSON: uptrend_analysis_YYYY-MM-DD_HHMMSS.json
+    - Markdown: uptrend_analysis_YYYY-MM-DD_HHMMSS.md
+"""
+
+import argparse
+import os
+import sys
+from datetime import datetime
+
+# Add parent directory to path for imports
+sys.path.insert(0, os.path.dirname(__file__))
+
+from data_fetcher import UptrendDataFetcher
+from calculators.market_breadth_calculator import calculate_market_breadth
+from calculators.sector_participation_calculator import calculate_sector_participation
+from calculators.sector_rotation_calculator import calculate_sector_rotation
+from calculators.momentum_calculator import calculate_momentum
+from calculators.historical_context_calculator import calculate_historical_context
+from scorer import calculate_composite_score
+from report_generator import generate_json_report, generate_markdown_report
+
+
+def parse_arguments():
+    parser = argparse.ArgumentParser(
+        description="Uptrend Analyzer - Market Breadth Health Diagnosis"
+    )
+    parser.add_argument(
+        "--output-dir",
+        default=".",
+        help="Output directory for reports (default: current directory)"
+    )
+    return parser.parse_args()
+
+
+def main():
+    args = parse_arguments()
+
+    print("=" * 70)
+    print("Uptrend Analyzer")
+    print("Market Breadth Health Diagnosis via Monty's Uptrend Ratio Dashboard")
+    print("=" * 70)
+    print()
+
+    # ========================================================================
+    # Step 1: Fetch CSV Data
+    # ========================================================================
+    print("Step 1: Fetching CSV Data")
+    print("-" * 70)
+
+    fetcher = UptrendDataFetcher()
+
+    print("  Fetching timeseries data...", end=" ", flush=True)
+    timeseries = fetcher.fetch_timeseries()
+    if timeseries:
+        print(f"OK ({len(timeseries)} rows)")
+    else:
+        print("FAILED")
+        print("ERROR: Cannot proceed without timeseries data", file=sys.stderr)
+        sys.exit(1)
+
+    print("  Fetching sector summary...", end=" ", flush=True)
+    sector_summary = fetcher.fetch_sector_summary()
+    if sector_summary:
+        print(f"OK ({len(sector_summary)} sectors)")
+    else:
+        print("WARN - Sector summary unavailable, some components will use defaults")
+
+    # Extract key data subsets
+    all_timeseries = fetcher.get_all_timeseries()
+    latest_all = fetcher.get_latest_all()
+    sector_latest = fetcher.get_all_sector_latest()
+
+    if latest_all:
+        ratio_pct = round(latest_all["ratio"] * 100, 1) if latest_all.get("ratio") is not None else "N/A"
+        print(f"  Latest data: {latest_all.get('date', 'N/A')}, "
+              f"ratio={ratio_pct}%, trend={latest_all.get('trend', 'N/A')}")
+    print()
+
+    # ========================================================================
+    # Step 2: Calculate Components
+    # ========================================================================
+    print("Step 2: Calculating Components")
+    print("-" * 70)
+
+    # Component 1: Market Breadth (30%)
+    print("  [1/5] Market Breadth (Overall)...", end=" ", flush=True)
+    comp1 = calculate_market_breadth(latest_all, all_timeseries)
+    print(f"Score: {comp1['score']} ({comp1['signal']})")
+
+    # Component 2: Sector Participation (25%)
+    print("  [2/5] Sector Participation...", end=" ", flush=True)
+    comp2 = calculate_sector_participation(sector_summary, sector_latest)
+    print(f"Score: {comp2['score']} ({comp2['signal']})")
+
+    # Component 3: Sector Rotation (15%)
+    print("  [3/5] Sector Rotation...", end=" ", flush=True)
+    comp3 = calculate_sector_rotation(sector_summary, sector_latest)
+    print(f"Score: {comp3['score']} ({comp3['signal']})")
+
+    # Component 4: Momentum (20%)
+    print("  [4/5] Momentum...", end=" ", flush=True)
+    comp4 = calculate_momentum(all_timeseries, sector_summary)
+    print(f"Score: {comp4['score']} ({comp4['signal']})")
+
+    # Component 5: Historical Context (10%)
+    print("  [5/5] Historical Context...", end=" ", flush=True)
+    comp5 = calculate_historical_context(all_timeseries)
+    print(f"Score: {comp5['score']} ({comp5['signal']})")
+
+    print()
+
+    # ========================================================================
+    # Step 3: Composite Score
+    # ========================================================================
+    print("Step 3: Calculating Composite Score")
+    print("-" * 70)
+
+    component_scores = {
+        "market_breadth": comp1["score"],
+        "sector_participation": comp2["score"],
+        "sector_rotation": comp3["score"],
+        "momentum": comp4["score"],
+        "historical_context": comp5["score"],
+    }
+
+    data_availability = {
+        "market_breadth": comp1.get("data_available", True),
+        "sector_participation": comp2.get("data_available", True),
+        "sector_rotation": comp3.get("data_available", True),
+        "momentum": comp4.get("data_available", True),
+        "historical_context": comp5.get("data_available", True),
+    }
+
+    # Collect warning flags from components
+    warning_flags = {
+        "late_cycle": comp3.get("late_cycle_flag", False),
+        "high_spread": (comp2.get("spread") is not None and comp2["spread"] > 0.40),
+    }
+
+    composite = calculate_composite_score(component_scores, data_availability,
+                                          warning_flags)
+
+    print(f"  Composite Score: {composite['composite_score']}/100")
+    print(f"  Zone: {composite['zone']}")
+    print(f"  Exposure Guidance: {composite['exposure_guidance']}")
+    print(f"  Strongest: {composite['strongest_component']['label']} "
+          f"({composite['strongest_component']['score']})")
+    print(f"  Weakest: {composite['weakest_component']['label']} "
+          f"({composite['weakest_component']['score']})")
+    print()
+
+    # ========================================================================
+    # Step 4: Generate Reports
+    # ========================================================================
+    print("Step 4: Generating Reports")
+    print("-" * 70)
+
+    os.makedirs(args.output_dir, exist_ok=True)
+
+    # Build full analysis
+    analysis = {
+        "metadata": {
+            "generated_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            "data_source": "Monty's Uptrend Ratio Dashboard (GitHub CSV)",
+            "api_key_required": False,
+            "latest_data_date": latest_all.get("date", "N/A") if latest_all else "N/A",
+            "timeseries_rows": len(timeseries),
+            "sectors_available": len(sector_summary),
+        },
+        "composite": composite,
+        "components": {
+            "market_breadth": comp1,
+            "sector_participation": comp2,
+            "sector_rotation": comp3,
+            "momentum": comp4,
+            "historical_context": comp5,
+        },
+    }
+
+    timestamp = datetime.now().strftime("%Y-%m-%d_%H%M%S")
+    json_file = os.path.join(args.output_dir, f"uptrend_analysis_{timestamp}.json")
+    md_file = os.path.join(args.output_dir, f"uptrend_analysis_{timestamp}.md")
+
+    generate_json_report(analysis, json_file)
+    generate_markdown_report(analysis, md_file)
+
+    print()
+    print("=" * 70)
+    print("Uptrend Analysis Complete")
+    print("=" * 70)
+    print(f"  Composite Score: {composite['composite_score']}/100")
+    print(f"  Zone: {composite['zone']}")
+    print(f"  Exposure Guidance: {composite['exposure_guidance']}")
+    print(f"  JSON Report: {json_file}")
+    print(f"  Markdown Report: {md_file}")
+    print()
+
+
+if __name__ == "__main__":
+    main()

package/.claude/skills/us-market-bubble-detector/CHANGELOG.md

@@ -0,0 +1,118 @@
+# US Market Bubble Detector - Changelog
+
+## Version 2.1 (November 3, 2025)
+
+### Critical Issue Fixed
+**Problem Identified:** v2.0 allowed excessive qualitative adjustments based on unmeasured "narratives" and subjective impressions, leading to inflated bubble risk scores.
+
+**Example Case (Nov 3, 2025):**
+- Quantitative Score (Phase 2): 9 points (objective, data-driven)
+- Qualitative Adjustment (v2.0): +2 points
+  - Media Narrative: +1 (based on "elevated AI narrative" - NO DATA)
+  - Valuation: +1 (P/E 30.8 - DOUBLE COUNTED, ignored fundamental backing)
+- **Result**: 11/16 points → Euphoria phase → 40% risk budget (overly defensive)
+
+**Root Cause:** Confirmation bias - analyst had bearish conclusion first, then adjusted qualitative points to match expectation.
+
+### Changes in v2.1
+
+#### 1. Stricter Qualitative Criteria (MAX +3, down from +5)
+
+**A. Social Penetration (0-1 points)**
+- **v2.0**: Loose criteria, "general awareness" acceptable
+- **v2.1**: ALL three required:
+  - Direct user report of non-investor recommendations
+  - Specific examples with dates/names
+  - Multiple independent sources (minimum 3)
+
+**B. Media/Search Trends (0-1 points)**
+- **v2.0**: Subjective "many reports" acceptable
+- **v2.1**: BOTH required:
+  - Google Trends 5x+ YoY (measured data)
+  - Mainstream coverage confirmed (Time covers, TV specials with dates)
+- **Critical**: "Elevated narrative" without data = 0 points
+
+**C. Valuation Disconnect (0-1 points)**
+- **v2.0**: P/E >25 alone sufficient
+- **v2.1**: ALL required AND avoid double-counting:
+  - P/E >25 (if NOT in Phase 2)
+  - Fundamentals explicitly ignored in discourse
+  - "This time is different" documented in major media
+- **Self-check**: If companies have real earnings supporting valuations → 0 points
+
+#### 2. Confirmation Bias Prevention
+
+New mandatory checklist before adding ANY qualitative points:
+```
+□ Do I have concrete, measurable data? (not impressions)
+□ Would an independent observer reach the same conclusion?
+□ Am I avoiding double-counting with Phase 2 scores?
+□ Have I documented specific evidence with sources?
+```
+
+#### 3. Granular Risk Phases
+
+**New "Elevated Risk" Phase (8-9 points)**
+- **v2.0**: 9 points = Euphoria = 40% risk budget (extreme defensive)
+- **v2.1**: 9 points = Elevated Risk = 50-70% risk budget (balanced caution)
+
+**Updated Risk Budget Matrix:**
+| Score | Phase | v2.0 Risk Budget | v2.1 Risk Budget | Change |
+|-------|-------|------------------|------------------|--------|
+| 0-4 | Normal | 100% | 100% | - |
+| 5-7 | Caution | 70% | 70-80% | More flexible |
+| 8-9 | Elevated Risk | 40% (Euphoria) | 50-70% | **NEW PHASE** |
+| 10-12 | Euphoria | 40% | 40-50% | More balanced |
+| 13-15 | Critical | 20% | 20-30% | Reduced max |
+
+#### 4. Maximum Score Reduction
+
+- **v2.0**: 0-16 points (Phase 2: 12, Phase 3: -1 to +5)
+- **v2.1**: 0-15 points (Phase 2: 12, Phase 3: 0 to +3)
+
+### Impact on Nov 3, 2025 Analysis
+
+**Under v2.0:**
+- Score: 11/16 → Euphoria phase
+- Risk Budget: 40%
+- Positioning: Extreme defensive
+
+**Under v2.1 (corrected):**
+- Quantitative: 9/12 (unchanged, data-driven)
+- Qualitative:
+  - Media Narrative: 0 points (no Google Trends data)
+  - Valuation: 0 points (AI has fundamental backing, double-counting)
+- **Score: 9/15 → Elevated Risk phase**
+- **Risk Budget: 50-70%**
+- **Positioning: Cautious but not extreme**
+
+### Key Learnings
+
+1. **Data > Impressions**: "Elevated narrative" is not measurable evidence
+2. **Avoid Double-Counting**: Valuation in Phase 2 quantitative ≠ add again in Phase 3
+3. **Check Internal Consistency**: If report admits "AI has fundamental backing," then valuation disconnect score must be 0
+4. **Independent Verification**: All qualitative points must be verifiable by independent observers
+
+### Documentation Updates
+
+- `SKILL.md`: Updated to v2.1 with strict criteria
+- `references/implementation_guide.md`: Enhanced Phase 3 with bias prevention checklist
+- `references/quick_reference.md`: Updated action matrix with new Elevated Risk phase
+- `references/bubble_framework.md`: Updated risk budget table
+
+---
+
+## Version 2.0 (October 27, 2025)
+
+### Initial Major Revision
+- Introduced mandatory quantitative data collection
+- Eliminated reliance on impressions and speculation
+- Established clear threshold settings for each indicator
+- Two-phase evaluation process: Quantitative → Qualitative
+
+---
+
+**Version Control:**
+- v1.x: Original framework (deprecated)
+- v2.0: Data-driven quantitative focus
+- v2.1: Strict qualitative criteria + confirmation bias prevention