feat: 50-signal macro engine — vol surface, sectors, EM, carry, long bonds
data_fetcher.py
- MACRO_GAUGE_CONFIG: 15 → 29 tickers (+silver, vvix, skew, ovx, gvz,
usdjpy, xlk, xlf, xlp, xlu, eem, emb, fxi, tlt)
- 5 new derived metrics: silver_gold_ratio, xlk_xlp_momentum, xlf_spx_ratio,
eem_spx_ratio, vol_surface_regime (composite classification)
- ThreadPoolExecutor max_workers raised to 20
- score_macro_scenarios: +15 new variables; each of 8 scenarios enriched
with vol-surface (SKEW, VVIX), sector rotation (XLK, XLF, XLP, XLU),
EM/carry (EEM, EMB, USDJPY), long bonds (TLT), silver signals
ai_analyzer.py
- macro_ctx: 5 → 21 fields per pattern (vol surface, sectors, EM, carry,
long bonds, silver/gold ratio — all with interpretation comments)
- macro_section in scoring prompt: describes surface de vol regime, sector
rotation, global/carry signals with explicit GPT instructions for pilier 3e
- DEFAULT_ANALYSIS_TEMPLATE: pilier 3e expanded with SKEW/VVIX/OVX/GVZ guidance
SIGNALS_FUTURES.md: reference document listing 30+ signals not yet
available (FRED, CFTC COT, EIA, Baltic Dry, LME, credit spreads,
hedge fund positioning, central bank balance sheets) with implementation
priority and cost estimate.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
d8c0334feb