Commit Graph

7 Commits

Author SHA1 Message Date
OpenSquared
a31f6e21b1 docs: v6.1 — Specialist Desks v2 (COT, Forward Curves, Surprise, Hawk/Dove)
Section 12 refonte complète : 4 nouvelles sous-sections documentant le
COT CFTC (19 marchés par codes contrat stables), les Forward Curves
yfinance (contango/backwardation), le Surprise Index (actual-consensus)
et le Hawk/Dove Text Scorer GPT-4o-mini. Architecture : +2 tables DB
(cot_data, forward_curve_data), +6 colonnes specialist_reports. Schéma
et étape 4b mis à jour avec les nouvelles sources injectées dans le
contexte IA.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-23 18:59:24 +02:00
OpenSquared
3b7fa35456 feat: Specialist Desks v2 — COT, Forward Curves, Surprise Index, Hawk/Dove scorer
- COT Positioning: CFTC disaggregated + financial futures (19 markets) via Socrata free API
  net MM position % OI + weekly change stored in cot_data table
- Forward Curves: yfinance front-month vs +3M slope (8 commodities)
  contango/backwardation/flat stored in forward_curve_data table
- Surprise Index: consensus_estimate + actual_value on specialist_reports
  auto-computes surprise_score = actual - consensus on save
- Hawk/Dove Text Scorer: GPT-4o-mini endpoint for CB statements
  score -1..+1, label, summary, key_phrases (forex/bonds: hawk/dove; commodities: bull/bear)
- AI context injection: COT net positioning, forward curve structure,
  surprise scores, upcoming consensus estimates injected into all desk blocks
- Frontend: COT panel (net% bars), Forward Curves panel, SurpriseInput
  on report cards, Hawk/Dove scorer in forex/bonds config tab
- auto_cycle.py: non-blocking COT + curve refresh before each cycle

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-23 18:00:46 +02:00
OpenSquared
4423ad91db docs: update documentation to v5.0 — Phase 5, 19 pages, IV Gate, Pattern Convergence
Major additions since v4.4:
- Cover/version bump to v5.0 (5 phases, 19 pages)
- Section 1.4: pages table updated 15→19 (VaR Analysis, Position History, Backtest, Calendar, System Logs)
- Step 1 macro engine: 50 signals table (vol surface, sector rotation XLK/XLF/XLP/XLU, EM/carry EEM/EMB/USDJPY, silver, 5 derived metrics)
- New Section 2b: Cycle Context Engine (delta temporal, news decay, context snapshot, FRED, technical indicators by horizon, price discovery, replay, weekend-aware scheduler)
- Section 3: IV Gate (2-level threshold, formula, rationale) + dynamic IV Watchlist (3 types, bootstrap)
- New Section 5b: Pattern Convergence (8 categories, signal_direction, conviction_score, Trade Ideas tab, IBKR ticket auto-calc)
- Section 8: 6 new tables (system_logs, iv_watchlist, context_snapshots, iv_history, pnl_snapshots, var_snapshots) + new columns in existing tables
- Section 9: Journal updated to 7 tabs, guide pages 14–19 (VaR, Position History, Backtest, Calendar, System Logs, Config v5)
- Section 10: 5 new design decisions (IV Gate, watchlist DB, weekend scheduler, indicators by horizon, context snapshot)
- Section 11: 14 new glossary terms (cycle_meta, news decay, IV Gate, signal direction, conviction score, IBKR ticket, DTE, vol surface regime, watchlist dynamique, etc.)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-22 09:40:46 +02:00
OpenSquared
58c3767a9d feat: simulation portfolio surveillance + patterns grid/filter UI
Portfolio Monitor (v4.4):
- New portfolio_risk.py service: concentration by asset_class, directional
  conflict detection (same underlying, opposite directions), overweight alerts
- AI agent (Step 7b) runs GPT-4o-mini after each cycle log: assessment +
  prioritized actions + rebalance suggestion, persisted in system_logs
- GET /api/journal/portfolio-risk — full risk breakdown + latest AI monitor reco
- POST /api/journal/trade-check — pre-entry conflict & concentration check
- asset_class column added to trade_entry_prices (auto-migration + populated at INSERT)
- Journal: new "Risque Sim." tab with concentration bars, conflict alerts,
  AI recommendations; red badge on tab when danger alerts exist

PatternEditor:
- Grid view default (2-3 cols responsive), list toggle
- Asset class filter chips (energy/metals/agri/equities/indices/forex/rates)
- Sort: Date (default) / Score IA / Prob.
- Period filter: Tout / 7j / 30j
- Result count badge when filters active

Doc: v4.3 → v4.4, updated Journal/PatternEditor/cycle steps/schema/glossary

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 15:49:46 +02:00
OpenSquared
8122bd019c docs: v4.3 — trade lifecycle, exit defaults, PostgreSQL migration note
- Page 8 Journal de Bord: réécriture complète pour les 5 onglets, badges
  🎯/🛑, barre Cible/Stop, modal de fermeture, onglet Fermés avec stats
- Table des pages: Page 8 description mise à jour (Ouverts/Fermés/1-clic),
  Page 14 Config mention exit defaults, 15+ tables architecture
- Section 10 Décisions: 2 nouvelles entrées (3 modes de fermeture,
  confirmation humaine toujours requise), note migration PostgreSQL
- Glossaire: 4 nouvelles entrées (Cycle de vie, Exit Defaults,
  P&L Réalisé vs Théorique, Alert Type)
- Footer et titre: v4.2 → v4.3

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 14:23:34 +02:00
OpenSquared
0ba73cae0b docs: v4.2 — AI news enrichment pipeline + directional alignment documented
Updates Étape 2 (RSS pipeline) to describe the 3-layer architecture: deterministic
classification → GPT-4o-mini batch enrichment (impact, ai_dir_*, ai_resolution,
ai_insight) → _compute_ai_alignment() bonus injection in pattern scoring.

Rewrites Prompt 1 section with new batch scoring prompt structure and adds a
concrete counter-signal example (Gaza ceasefire → oil bearish, pattern downranked).

Updates Radar Géopolitique usage guide with AI badge reading guide and decision
rules (green badge + relevance ≥60% = candidate, red badge = wait for confirmation).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 13:34:33 +02:00
OpenSquared
44ada3f8f1 feat: stable macro regime detection — blended signals + Bayesian smoothing
Replace 1-day change_pct with weighted blend (20% 1d / 50% 5d / 30% 10d)
for all 20 scored signals. Add Bayesian prior from rolling 25-day history
(weight 15%→45%) and a 10-point persistence threshold before regime switch.

Bootstrap on first load: replays last 20 trading days via yf.download batch
(45d) to pre-populate _regime_history, so stability is visible immediately.

Frontend: adds 'stable Xj' badge and history depth indicator on regime banner.
Doc: updates v4.0→v4.1, rewrites Étape 1 Régime Macro and glossary entry.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-19 12:40:38 +02:00