market-regime-analysis

Classify and explain the current market regime across trend, volatility, breadth, liquidity, rates, inflation, FX, credit, earnings revisions, positioning, correlations, and dated policy/event risk, t

SKILL.md

GitHub

Market Regime Analysis

Own the cross-asset environment, not the final security recommendation. Use current primary data and distinguish measurement from interpretation.

Read references/regime-protocol.md. Use scripts/regime_features.py for transparent price/volatility/breadth features when compatible dated data exists.

Workflow

  1. Define scope. Establish geography, asset universe, currency, horizon and data cutoff. A daily trading regime and a multi-year macro regime are different questions.
  2. Collect current measures. Use official central-bank/statistics data for rates, inflation, labor and activity; exchange/primary market data for prices, breadth and liquidity; dated filings/estimates for earnings and credit when relevant.
  3. Measure independent axes. Report trend, realized/implied volatility, breadth, liquidity/funding, rates/yield curve, inflation/growth direction, FX, credit, earnings revisions and event calendar separately before assigning a label.
  4. Check disagreement. Highlight mixed regimes such as index strength with weak breadth, falling inflation with tightening liquidity, or calm volatility with concentrated event risk.
  5. Compare history carefully. Use matched historical episodes as base rates, not templates. State sample, differences, data vintage and whether the current regime is outside the observed range.
  6. Map transmission. Explain which assets, sectors, durations, currencies, factors and strategies are helped or hurt, and through what mechanism. Separate first-order from second-order effects.
  7. Build transitions. Give base, improvement and deterioration paths with observable triggers, not a permanent one-word label.
  8. Route action. Send instrument timing to $technical-quant-analysis, outcome ranges to $probabilistic-market-forecast, portfolio exposure to $portfolio-risk-and-sizing, and final decisions to $investment-copilot.

Hard rules

  • Do not reduce the regime to one index, indicator, headline or “risk-on/off” label.
  • Do not use revised macro data as if it was known in historical real time.
  • Do not infer causality from a recent correlation.
  • Do not assume correlations, liquidity or policy transmission stay stable under stress.
  • Do not issue a naked BUY/SELL solely from regime classification.

Output

Lead with scope, cutoff and a one-sentence regime assessment. Then show the axis dashboard, conflicts, historical base-rate limits, asset/strategy transmission, transition triggers, key dated events, implications for the requested horizon and confidence.