Superior-Trade/trading-terminal
Self-hosted AI trading terminal for Hyperliquid: chart read in, validated Freqtrade/Nautilus-Trader strategy out, deployed live.
What it solves
It bridges the gap between a trader's visual intuition (drawing on a chart) and the execution of a programmatic trading strategy. It eliminates the manual effort of translating a visual trade setup into a coded strategy and provides a safety layer to prevent common coding errors that lead to financial loss.
How it works
Users draw their market predictions (trendlines, levels, zones) directly on a chart. An AI agent analyzes the drawings, the visible chart range, and a screenshot of the chart to propose a trade plan with specific entry, stop, and target levels. Once the user approves, the agent generates Freqtrade strategy code in Python. This code is then passed through a deterministic validator that checks for critical errors (such as "look-ahead bias" or incorrect leverage scaling) and automatically asks the agent to repair the code until it passes all safety checks before deployment.
Who it’s for
Traders using Hyperliquid who want to automate their visual analysis and move from a "sketch" to a live, validated trading bot without writing the code themselves.
Highlights
- Visual-to-Code Pipeline: Converts freehand chart drawings and screenshots into executable Python strategies.
- Deterministic Safety Validator: A repair loop that catches common algorithmic trading bugs (e.g., reading future candles or NaN indicators) before they reach production.
- Integrated Execution: Supports live deployments, bracket orders, position sizing, and leverage caps.
- Flexible Deployment: Available as a hosted service or a self-hosted version where users maintain control of their API keys and prompts.
Related
- Project
- Project
- Project
- Project