Self-tuning multi-agent AI trading system with a public calibration log 8-source signal fusion · 10 ML models incl. Kronos · Bull/Bear/Judge debate on Claude Opus 4.7 Polymarket + Kalshi prediction markets vote alongside ML models. Weights adapt to per-source accuracy automatically. Different from other AI-trading repos. Orallexa isn't a hedge-fund simulator. It's a calibrated multi-agent decision system with a published Brier log, a FinPos-style Risk Sizer that keeps direction and position sizing on separate agents, and a hard kill switch. Paper-only. Real money is gated behind a written Brier improvement, not vibes. Live Demo · Presentation · Evaluation Report · 中文 What makes this different Most AI trading projects: feed data into a model, get a signal, done. Orallexa runs a multi-agent intelligence pipeline. 4 AI analysts with different risk profiles debate the trade. A 20-agent swarm simulates market reactions. 5 independent signal sources vote. A bias tracker corrects the system's own mistakes. Then it executes. Every stage automated. Every stage observable. The system learns fr

⚠ No verified equity curve — no track-record source connected.

No score history is stored yet — only the current score is shown.

⚠ No live verification account connected — ask for proof before buying.

Alerts on changes: coming soon

Prop-firm compatibility not provided.

Data-completeness & trust index (not a profitability rating)