Appearance

Chat Interface Guide ​

This guide walks you through using Robonet's chat interface to build, test, and deploy algorithmic trading strategies through natural conversation.

Overview ​

The chat interface provides a conversational way to:

  • Generate trading strategy ideas
  • Implement strategies with AI assistance
  • Run backtests and analyze results
  • Optimize strategy parameters
  • Integrate Allora Network price predictions
  • Deploy strategies to live trading

For traders who prefer natural language over code.

Getting Started ​

1. Account Creation and Login ​

Creating Your Account ​

  1. Visit the Robonet platform and click "Sign In"
  2. You'll be prompted to authenticate via Privy
  3. Privy creates an embedded wallet for you automatically
  4. If you have an invitation access code, enter it when prompted
  5. Complete authentication to access your account

Wallet Setup ​

After authentication, Robonet uses Privy for secure wallet management:

  1. Embedded Wallet: Privy automatically creates a secure embedded wallet for you
  2. Wallet Delegation: Authorize Robonet to sign trades on your behalf
    • This is a one-time setup step
    • Required for live trading deployments
    • You can revoke delegation at any time
  3. External Wallets: You can also connect external Ethereum wallets if preferred

Important: Wallet delegation only allows trading operations. Robonet cannot withdraw funds from your wallet.

Network Information ​

  • Trading: All trades execute on Hyperliquid Perpetuals (Hyperliquid L1)
  • Payments: Credits are managed on Base network (USDC)
  • Settlement: All trades settle on-chain on Hyperliquid
  • Your chat sessions and strategies are linked to your wallet address

2. Starting a Chat Session ​

Once logged in:

  1. Click "New Chat" in the sidebar to start a fresh session
  2. Previous sessions are saved and accessible from the sidebar
  3. Each session maintains its own conversation history
  4. Sessions auto-save to your browser's local storage

Screenshot placeholder: Chat interface with sidebar showing session list

Building Strategies ​

Generating Strategy Ideas ​

The chat interface helps you brainstorm and refine trading ideas.

Example conversation:

You: I want to build a momentum strategy for Bitcoin
AI: I can help you create a momentum-based strategy. Let me suggest a few approaches:

1. RSI + Moving Average Crossover
2. Breakout strategy with volume confirmation
3. Trend-following with ATR stops

Which approach interests you most, or would you like to explore a different direction?

You: Let's go with the RSI and moving average approach

AI: Great choice! I'll create a momentum strategy that:
- Uses RSI to identify oversold/overbought conditions
- Confirms with EMA crossovers for trend direction
- Includes ATR-based stop losses for risk management

Let me implement this strategy for you...

The AI will:

  1. Ask clarifying questions about your preferences
  2. Suggest multiple approaches
  3. Explain the reasoning behind each approach
  4. Generate a complete strategy implementation

Requesting Strategy Implementation ​

Once you've refined your idea, ask the AI to implement it:

You: Create this strategy for BTC-USDT on the 4-hour timeframe

What happens behind the scenes:

  1. The AI calls the create_strategy tool
  2. You'll see a tool execution status card showing progress
  3. The strategy code is generated based on your conversation
  4. The completed strategy is saved to your account

Tool Execution Interface:

While tools execute, you'll see:

  • Tool name and status (running/success/error)
  • Progress indicators with estimated time
  • Expandable details showing execution steps
  • Results or error messages when complete

Screenshot placeholder: Tool execution status card showing running state

Testing Strategies ​

Running Backtests ​

Once a strategy is created, test it with historical data:

You: Backtest MomentumRSI on BTC-USDT with 4h timeframe for the last 6 months

The platform will:

  1. Execute the run_backtest tool
  2. Run your strategy against historical data
  3. Generate performance metrics and equity curve
  4. Display results in an interactive card

Backtest Parameters:

  • Symbol: Trading pair (e.g., BTC-USDT)
  • Timeframe: Candlestick interval (1m, 5m, 15m, 30m, 1h, 2h, 4h, 6h, 12h, 1d)
  • Date Range: Historical period to test (default: 6 months)
  • Data Source: Currently Hyperliquid Perpetual only

Interpreting Backtest Results ​

The backtest result card displays key metrics:

Performance Metrics ​

  • Total Return: Overall percentage gain/loss
  • Net Profit: Dollar amount of profit/loss
  • Annual Return: Annualized performance
  • Sharpe Ratio: Risk-adjusted return (higher is better)
    • < 1.0: Poor risk-adjusted returns
    • 1.0 - 2.0: Good performance
    • > 2.0: Excellent performance
  • Sortino Ratio: Similar to Sharpe, but only penalizes downside volatility

Risk Metrics ​

  • Max Drawdown: Largest peak-to-trough decline (lower is better)
    • < 20%: Conservative
    • 20-40%: Moderate risk
    • > 40%: Aggressive/high risk
  • Calmar Ratio: Return divided by max drawdown (higher is better)

Trading Activity ​

  • Total Trades: Number of completed trades
  • Win Rate: Percentage of profitable trades
    • < 40%: Needs improvement
    • 40-60%: Typical for many strategies
    • > 60%: Strong performance
  • Profit Factor: Gross profit / gross loss (above 1.0 = profitable)
  • Avg Win/Loss: Average profit vs loss per trade

Equity Curve ​

The equity curve shows your account value over time:

  • Upward trend: Strategy is profitable
  • Smooth curve: Consistent performance
  • Volatile swings: High risk, consider optimization
  • Flat line: Strategy isn't trading or is break-even

Screenshot placeholder: Backtest result card with metrics and equity curve chart

Understanding "No Trades" Results ​

If your backtest shows "No trades executed":

Common causes:

  • Entry conditions are too strict
  • Timeframe doesn't match strategy logic
  • Insufficient historical data for indicators
  • Symbol doesn't match strategy requirements

What to do:

  1. Review strategy entry/exit conditions
  2. Try a longer backtest period
  3. Adjust indicator parameters
  4. Ask the AI to optimize the strategy

Optimizing Strategies ​

Strategy Optimization Process ​

If backtest results aren't satisfactory, optimize your strategy:

You: Optimize MomentumRSI to maximize Sharpe ratio

The AI will:

  1. Call the optimize_strategy tool
  2. Test multiple parameter combinations
  3. Find optimal values for:
    • Indicator periods (RSI, EMA, ATR, etc.)
    • Entry/exit thresholds
    • Risk management parameters
  4. Return improved parameters with performance comparison

Optimization Targets:

  • Sharpe Ratio: Maximize risk-adjusted returns
  • Total Return: Maximize absolute profit
  • Max Drawdown: Minimize largest loss
  • Win Rate: Maximize percentage of winning trades
  • Profit Factor: Maximize profit-to-loss ratio

Example result:

Original Strategy:
- RSI Period: 14
- RSI Oversold: 30
- RSI Overbought: 70
- Sharpe Ratio: 1.2

Optimized Strategy:
- RSI Period: 21
- RSI Oversold: 25
- RSI Overbought: 75
- Sharpe Ratio: 2.1 (+75% improvement)

Important Considerations:

  • Overfitting Risk: Optimized strategies may perform worse in live trading if over-tuned to historical data
  • Multiple Backtests: Test optimized parameters across different time periods
  • Walk-Forward Analysis: Ask the AI to validate on out-of-sample data

Allora Network Integration ​

Price Prediction Enhancement ​

Allora Network provides ML-powered price predictions. Integrate them into your strategy:

You: Enhance MomentumRSI with Allora Network price predictions

The AI will:

  1. Call the enhance_with_allora tool
  2. Integrate Allora predictions as additional signals
  3. Adjust entry/exit logic to incorporate ML forecasts
  4. Generate a new strategy version with Allora integration

How Allora Predictions Work:

  • Allora Network aggregates predictions from multiple AI models
  • Predictions are made for various time horizons (short/medium/long-term)
  • Your strategy can use predictions as:
    • Confirmation signals: Enter only when prediction agrees with technical indicators
    • Filters: Avoid trades against predicted direction
    • Weightings: Adjust position sizing based on prediction confidence

Example Integration:

Original Entry Condition:
- RSI < 30 (oversold)
- Fast EMA crosses above Slow EMA

Enhanced Entry Condition:
- RSI < 30 (oversold)
- Fast EMA crosses above Slow EMA
- Allora 24h prediction shows > 2% upside

Benefits:

  • Improved win rate by filtering false signals
  • Better risk management through prediction confidence
  • Access to institutional-grade ML models
  • Reduced drawdowns during market regime changes

Prediction Market Strategies ​

Create strategies specifically for prediction markets (Polymarket, etc.):

You: Create a prediction market strategy for BTC price in February 2026

The AI will:

  1. Call get_prediction_market_data to fetch market timeseries
  2. Analyze YES/NO token pricing trends
  3. Generate a strategy to trade prediction markets
  4. Backtest on historical prediction market data

Prediction Market Card:

View prediction market data directly in chat:

  • YES/NO token prices over time
  • Market resolution status
  • High/low/current prices
  • Interactive dual-token chart

Screenshot placeholder: Prediction market card with YES/NO token timeseries

Deploying Strategies ​

Deployment Overview ​

Once you're satisfied with backtest results, deploy your strategy to live trading.

Deployment Types ​

EOA (Externally Owned Account):

  • Trades directly with your connected wallet
  • Maximum 1 active deployment per user
  • Immediate setup, no minimum balance
  • Suited for: Testing with small capital

Hyperliquid Vault:

  • Creates a Hyperliquid Vault for your strategy
  • Unlimited vaults per user
  • Requires 200 USDC minimum
  • Suited for: Production trading with larger capital

Creating a Deployment ​

From chat, request deployment:

You: Deploy MomentumRSI to BTC-USDT on 4h timeframe with 2x leverage

Or navigate to your Profile → Strategies tab and click "Deploy" on a saved strategy.

Deployment Configuration:

  1. Deployment Type: EOA or Hyperliquid Vault
  2. Symbol: Trading pair (e.g., BTC-USDT)
  3. Timeframe: Candlestick interval (must match backtest)
  4. Leverage: 1x - 5x (higher leverage = higher risk)
  5. Vault Details (if Hyperliquid Vault deployment):
    • Vault Name (unique on Hyperliquid)
    • Description (minimum 10 characters)

Risk Warnings:

  • Leverage amplifies both gains and losses
  • Start with lower leverage (1-2x) until proven profitable
  • Monitor positions regularly
  • Set appropriate stop losses in your strategy

Deployment Lifecycle ​

Status Flow:

  1. Pending: Deployment request submitted
  2. Building: Docker container being created
  3. Starting: Kubernetes workload launching
  4. Running: Strategy is live and trading
  5. Stopping: Deployment shutdown in progress
  6. Stopped: Deployment inactive
  7. Failed: Error occurred (check logs)

Monitoring:

The Live Strategies page shows:

  • Status: Current deployment state
  • TVL: Total Value Locked (vault only)
  • PnL: Profit and Loss (absolute + percentage)
  • Returns: 24h/7d/30d performance
  • Chart: Mini equity curve
  • Trading: Active/inactive indicator
  • Actions: Stop/Start/Retry buttons

Optimistic UI:

When you stop/start a deployment, the UI updates immediately while the API call completes in the background. This provides instant feedback without waiting for server response.

Screenshot placeholder: Live Strategies table with multiple deployments

Managing Deployments ​

From the Chat Interface:

You: Stop my BTC-USDT deployment

You: Start the MomentumRSI deployment again

From the Live Strategies Page:

  1. Navigate to Profile → Live Strategies
  2. Find your deployment in the table
  3. Click the action button (⋮) in the Actions column
  4. Select Stop/Start/Retry

Deployment Constraints:

  • EOA Deployments: Only 1 active at a time (must stop existing before starting a new one)
  • Hyperliquid Vault Deployments: Unlimited active deployments
  • Wallet Delegation Required: Must authorize delegation before deploying
  • Insufficient Balance: Deployment will fail if insufficient USDC in wallet

Error Handling:

  • Insufficient Credits: Add credits to your account
  • Failed Deployment: Check error message and retry
  • Connection Issues: Deployments continue running even if you disconnect

Tips and Best Practices ​

Strategy Development ​

  1. Start Simple: Begin with basic strategies before adding complexity
  2. Backtest Thoroughly: Test on at least 6 months of data, preferably 1+ year
  3. Multiple Timeframes: Validate strategy on different timeframes
  4. Out-of-Sample Testing: Reserve recent data for final validation
  5. Walk-Forward: Test on rolling windows, not just one period

Risk Management ​

  1. Conservative Leverage: Start with 1-2x leverage
  2. Position Sizing: Never risk more than 2-5% of capital per trade
  3. Stop Losses: Always include stop losses in your strategy
  4. Max Drawdown: Don't deploy strategies with >40% historical drawdown
  5. Diversification: Run multiple uncorrelated strategies

Live Trading ​

  1. Paper Trade First: Use testnet to verify strategy execution
  2. Small Capital: Start with minimum viable amount
  3. Monitor Closely: Check deployments daily for the first week
  4. Performance Tracking: Compare live performance to backtest
  5. Kill Switches: Know how to stop deployments quickly if needed

Using the Chat Effectively ​

  1. Be Specific: Provide clear requirements (symbol, timeframe, risk tolerance)
  2. Iterate: Start with a concept, refine through conversation
  3. Ask Questions: The AI can explain concepts and strategy logic
  4. Request Modifications: Ask for specific changes to strategies
  5. Save Sessions: Name sessions descriptively for easy reference

Troubleshooting ​

Common Issues ​

"Insufficient Credits" Error:

  • Solution: Add credits via Profile → Billing
  • Tool execution requires credits based on complexity
  • AI tools bill actual LLM cost + margin

Deployment Won't Start:

  • Check wallet delegation is authorized
  • Verify sufficient USDC balance (200 min for vaults)
  • Ensure no other EOA deployment is active
  • Check error message in deployment status

No Trades in Backtest:

  • Entry conditions may be too restrictive
  • Try longer backtest period
  • Adjust indicator parameters
  • Ask AI to optimize or debug strategy

Chat Session Not Loading:

  • Clear browser cache and reload
  • Check network connection
  • Try creating a new session
  • Contact support if issue persists

Message Queue Stuck:

  • If message shows "sending" indefinitely, refresh page
  • Previous message must complete before next is sent
  • Cancel operation if needed

Getting Help ​

In-Chat Support:

You: I need help understanding why my strategy has no trades

The AI can:

  • Debug strategy logic
  • Explain metrics and results
  • Suggest improvements
  • Guide through platform features

External Resources:

Next Steps ​

Note: Screenshots will be added in a future update. Placeholder notes indicate where visual aids will be inserted.