atlas-prediction-markets

name	atlas-prediction-markets
description	Prediction market data flow, Polymarket integration, sentiment momentum. Use when adding prediction markets, extending Polymarket API, or wiring momentum into signals.

Prediction Markets Integration

Data Flow

1. api/osint/polymarket.ts — Gamma API, basic events (existing)
2. api/osint/polymarket-history.ts — CLOB prices-history for sparklines
3. api/osint/polymarket-metrics.ts — Aggregates Gamma + CLOB, computes sentimentMomentum
4. DataService — fetchPolymarketMetrics(), dispatches 'polymarket-metrics'
5. Panels — Subscribe to event, render probability bars + sparklines + momentum
6. Sentiment strategy — Reads momentum from state, boosts confidence when aligned with GDELT tone

Key Interfaces (shared or src/lib)

• PredictionMarket — id, probability, volume24h, category, endDate, outcomePrices
• MarketMetrics — marketId, probability, volume24h, sentimentMomentum, history24h
• HistoricalOdds — marketId, points: {t, p}[], interval

Sentiment Momentum Formula

momentum = (price_now - price_24h_ago) / max(price_24h_ago, 0.01)

Clamp to [-1, 1]. Positive = moving toward Yes, negative = toward No.

Polymarket APIs

• Gamma: gamma-api.polymarket.com/events — events, outcomePrices, volume
• CLOB: clob.polymarket.com/prices-history?market={id}&interval=1h — historical Yes prices
• Map Gamma market conditionId/clobTokenIds → CLOB market param

Integration Points

• State: state.set('predictionMarkets', metrics) when polymarket-metrics fires
• Sentiment strategy: Check momentum; if > 0.1 or < -0.1, modify confidence
• Panel: dataService.addEventListener('polymarket-metrics', ...)

Reference

• Full prompt strategy: docs/PREDICTION-MARKETS-PROMPT-STRATEGY.md