https://github.com/meiiie/hinto-trader
Local-first crypto futures trading engine with FastAPI, React, Tauri, backtesting, and paper-mode safety defaults.
https://github.com/meiiie/hinto-trader
algorithmic-trading backtesting binance-futures crypto fastapi hinto python react risk-management tauri trading-bot
Last synced: 15 days ago
JSON representation
Local-first crypto futures trading engine with FastAPI, React, Tauri, backtesting, and paper-mode safety defaults.
- Host: GitHub
- URL: https://github.com/meiiie/hinto-trader
- Owner: meiiie
- License: mit
- Created: 2026-05-10T07:31:36.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2026-05-22T16:18:10.000Z (about 1 month ago)
- Last Synced: 2026-05-22T16:45:04.288Z (about 1 month ago)
- Topics: algorithmic-trading, backtesting, binance-futures, crypto, fastapi, hinto, python, react, risk-management, tauri, trading-bot
- Language: Python
- Size: 5.81 MB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Security: SECURITY.md
Awesome Lists containing this project
README
Local-first crypto futures trading engine with backtesting, paper mode, live monitoring, and a Tauri desktop dashboard.
Features |
Screenshots |
Architecture |
Quick Start |
Safety |
Strategy Direction |
Contributing
---
## Overview
**Hinto** is a research-grade trading system for Binance Futures. It keeps the
core trading logic local, tracks signals away from the exchange, and executes
only when a tracked signal is triggered. The project includes:
- A FastAPI backend for market data, signal generation, execution, settings,
analytics, and WebSocket updates.
- A React + Tauri dashboard for monitoring signals, positions, backtests, and
live state.
- A backtesting engine tuned for closer live parity with 1m monitoring,
candle-close stop validation, and realistic execution assumptions.
- Guardrails for paper/testnet-first operation, secret handling, and public
release hygiene.
Hinto is open-sourced as a codebase and research platform. It is not financial
advice, a profit guarantee, or a plug-and-play money machine.
## Features
| Area | What Hinto Provides |
| --- | --- |
| Local-first execution | Signals are stored locally and only become orders when price triggers. |
| Strategy engine | Liquidity-zone style signal generation, confluence scoring, delta divergence, and multi-timeframe trend filters. |
| Risk controls | Candle-close stop loss, hard-cap protection, exchange backup stops, global drawdown halt, and dead-zone windows. |
| Backtesting | Historical simulation with 1m monitoring, configurable leverage, max positions, stop logic, and execution mode. |
| Dashboard | React/Tauri interface for charts, portfolio state, signal logs, settings, and backtest views. |
| Persistence | SQLite repositories for signals, orders, positions, settings, and analytics. |
| Notifications | Telegram notification support for system and trade events. |
| Public safety | `.env.example` templates, tracked-file secret scanning, ignored local secrets, and paper mode defaults. |
## Screenshots
## Architecture
```mermaid
flowchart LR
ws["Binance WebSocket"] --> shared["Shared Binance Client"]
shared --> signal["Signal Generator"]
signal --> shark["Shark Tank Coordinator"]
shark --> live["Live Trading Service"]
live --> monitor["Position Monitor"]
monitor --> rest["Binance REST API"]
live --> db["SQLite Repositories"]
live --> tg["Telegram Notifications"]
db --> api["FastAPI"]
api --> ui["React + Tauri Dashboard"]
```
The backend follows a layered structure:
| Layer | Path | Responsibility |
| --- | --- | --- |
| API | `backend/src/api` | HTTP/WebSocket endpoints and request schemas. |
| Application | `backend/src/application` | Trading, backtesting, signal, risk, and orchestration services. |
| Domain | `backend/src/domain` | Entities, interfaces, value objects, and repository contracts. |
| Infrastructure | `backend/src/infrastructure` | Binance clients, persistence, indicators, notifications, execution queues. |
| Frontend | `frontend/src` | Desktop dashboard, charts, settings, and monitoring views. |
## Quick Start
Clone and configure:
```powershell
git clone https://github.com/meiiie/hinto-trader.git
cd hinto-trader
Copy-Item .env.example .env
```
The default `.env.example` uses `ENV=paper` with `HINTO_PAPER_REAL=true`.
This keeps live Binance market data and a live-like dashboard while orders and
fills stay in the local simulator. Keep this mode while validating strategy
behavior; see `docs/PAPER_REAL_MODE.md`.
Run the backend:
```powershell
cd backend
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
python run_backend.py
```
Run the dashboard:
```powershell
cd frontend
npm ci
npm run dev
```
Build the dashboard:
```powershell
cd frontend
npm run build
```
Run the tracked-file secret scan:
```powershell
powershell -NoProfile -ExecutionPolicy Bypass -File .\scripts\secret-scan.ps1
```
## Configuration
Hinto reads runtime configuration from environment files and local settings.
Never commit real credentials.
| File | Purpose |
| --- | --- |
| `.env.example` | Root template for paper/testnet/live settings. |
| `backend/.env.example` | Backend-specific Binance/testnet template. |
| `frontend/.env.example` | Dashboard API/WebSocket endpoint template. |
| `SECURITY.md` | Secret-handling and reporting guidance. |
## Safety
Trading software can lose money quickly, especially with leverage. Treat Hinto
as research infrastructure unless you have audited the code, the configuration,
and the exchange behavior yourself.
- Start with `ENV=paper`.
- Use Binance testnet before live credentials.
- Rotate any key that has ever been copied into a repository, chat, screenshot,
or log.
- Do not commit `.env`, `.env.production`, PEM files, databases, generated CSVs,
or local logs.
- Run `scripts/secret-scan.ps1` before every public release.
## Strategy Direction
Hinto is being developed as a research platform before it is treated as a live
trading system. The current mean-reversion scalper is now a benchmark track, not
a final answer. The next research track focuses on positive-skew continuation:
small controlled losses, larger trailed winners, and validation by R-multiple
distribution rather than headline win rate.
See `docs/STRATEGY_ROADMAP.md` for promotion/kill criteria, broker expansion
rules, and the clean-architecture path for adding new strategies.
## Repository Status
This public repository starts from a cleaned source snapshot. Private deployment
history, local logs, old agent folders, generated build outputs, and archived
research scratch files were intentionally left out.
## Contributing
Contributions are welcome. Please read `CONTRIBUTING.md` before opening an
issue or pull request.
Good first areas:
- Safer defaults and clearer configuration validation.
- Backtest/live parity tests.
- Dashboard usability and accessibility.
- Documentation for paper/testnet workflows.
- Exchange adapter abstractions.
## License
MIT License. See `LICENSE`.