Platform Overview
Autonomous self-learning cryptocurrency paper trading terminal connected to public multi-exchange feeds and Supabase PostgreSQL memory with Row-Level Security (RLS).
Platform High-Impact Metrics
Zenth is engineered for quantitative signal discovery, execution simulation, and continuous failure learning without financial risk.
Core Value Proposition
Zenth operates in strict simulated paper trading mode (`mode: PAPER`), eliminating capital risk while evaluating quantitative momentum strategies against genuine, live cryptocurrency tick data across Binance, Coinbase, OKX, Upbit, Bitget, and XT.com.
CRITICAL SYSTEM INVARIANTS
1. Zero Private API Keys: Zenth NEVER requires or stores exchange secret keys. 2. Hard $1,000 Cap: Maximum allocation limit of $1,000.00 USD/USDT per trade. 3. Under 200 Lines/File: Strict modularity invariant across the entire codebase.
Autonomous System Pipeline
The trading lifecycle is structured as a decoupled multi-tier pipeline with continuous failure distillation.
| Subsystem Layer | Domain Package | Key Operational Responsibility |
|---|---|---|
| 1. Market Feed Ingestion | src/core/market | Direct public REST polling with interval and pair normalization |
| 2. Quantitative Strategy | src/core/strategy | SMA 9/21 crossover, RSI 14 momentum, Volume SMA 20 filter |
| 3. Risk Management | src/core/risk | $1,000 notional ceiling, daily loss & loss streak circuit breakers |
| 4. Execution Simulation | src/core/execution | Paper order simulator with Stop-Loss (1.5%) and Take-Profit (3.0%) |
| 5. Adaptive Learning | src/core/memory | Real-time failure ingestion & pre-trade pattern rejection |
| 6. Fullscreen TUI | src/tui | Touch & click terminal with pinned 3-row HUD & 14 color themes |
| 7. Multi-Format Exporter | src/core/export | Session exports to TXT, CSV, Markdown, Word DOCX, and PDF 1.4 |