MCP Server Setup

The Robonet MCP (Model Context Protocol) server allows you to integrate Robonet's trading strategy tools directly into your AI coding environment. This guide will help you set up the MCP server for use with Claude Code, Cursor, Windsurf, or other MCP-compatible clients.

Quick Start

Get started with Robonet MCP in under 5 minutes:

1. Get Your API Key

Sign up at robonet.finance
Navigate to Settings → API Keys → Generate New API Key
Copy your API key (store it securely - you won't see it again)

2. Configure Your Client

Choose your AI coding assistant and add the Robonet MCP server:

json

// Add to: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
// Or: %APPDATA%\Claude\claude_desktop_config.json (Windows)
{
  "mcpServers": {
    "robonet": {
      "type": "http",
      "url": "https://mcp.robonet.finance/mcp?api_key=<API_KEY>"
    }
  }
}

claude mcp add robonet --transport http https://mcp.robonet.finance/mcp?api_key=<API_KEY>

json

// Add to: ~/.claude.json (macOS/Linux)
// Or: <project folder>\.mcp.json (Windows)
{
  "mcpServers": {
    "robonet": {
      "type": "http",
      "url": "https://mcp.robonet.finance/mcp?api_key=<API_KEY>"
    }
  }
}

json

// Open Cursor Settings → MCP → Add MCP Server
{
  "mcpServers": {
    "robonet": {
      "type": "http",
      "url": "https://mcp.robonet.finance/mcp?api_key=<API_KEY>"
    }
  }
}

toml

# Add to your .codex/config.toml
[mcp_servers.robonet]
url = "https://mcp.robonet.finance/mcp?api_key=<API_KEY>"

WARNING

Replace <API_KEY> with your actual API key from Step 1. Never commit your API key to version control.

3. Verify Setup

Restart your AI client
Ask: "Can you list available trading symbols on Robonet?"
Your AI should call the get_all_symbols MCP tool and return symbols like ETH-USD, BTC-USD, etc.

If you see trading symbols, you're set. Try creating your first strategy:

"Generate a simple RSI mean reversion strategy for ETH-USD and backtest it"

Overview

The MCP server provides 24 specialized tools for building, testing, and deploying trading strategies:

Data tools (8): Browse strategies, symbols, indicators, Allora topics, and data availability
AI-powered tools (6): Generate ideas, create strategies, optimize, enhance, and refine
Backtesting (2): Run backtests on crypto and prediction markets
Prediction markets (3): Browse events and market data
Deployment (4): Create, list, start, and stop live trading agents
Account (2): Check credit balance and transaction history

Example Workflow

Here's a typical workflow using the Robonet MCP server:

1. Generate Strategy Ideas

You: "Generate 3 trading strategy ideas based on current market conditions for ETH-USD"

AI: [Calls generate_ideas tool]
- Returns: 3 strategy concepts with entry/exit logic, risk management, and rationale

2. Create a Strategy

You: "Implement the RSI mean reversion strategy for ETH-USD on 1-hour timeframe"

AI: [Calls create_strategy tool]
- Returns: Complete Python strategy code following Jesse framework conventions

3. Run a Backtest

You: "Backtest this strategy from 2024-06-01 to 2025-01-01"

AI: [Calls run_backtest tool]
- Returns: Performance metrics (Sharpe ratio, win rate, max drawdown, total return, etc.)
- Includes: Equity curve data and trade statistics

4. Optimize Parameters

You: "Optimize the RSI period parameter between 10 and 20"

AI: [Calls optimize_strategy tool]
- Returns: Best parameter values and performance improvements
- Creates: New optimized version of the strategy

5. Enhance with Allora

You: "Enhance this strategy with Allora Network price predictions"

AI: [Calls enhance_with_allora tool]
- Returns: New strategy version that incorporates ML predictions as signals
- Preserves: Original strategy logic while adding predictive layer

6. Deploy to Production

You: "Deploy the RSI mean reversion strategy"

AI: [Calls deployment_create tool]
- Returns: Deployment status

Available MCP Tools

For a complete reference of all 17 MCP tools with detailed parameters and examples, see:

→ MCP Tools Reference

Troubleshooting

"Insufficient Credits" Error

Solution: Purchase credits in your Robonet dashboard:

Log in to robonet.finance
Navigate to Settings → Billing
Click Purchase Credits
Complete the payment

"Invalid API Key" Error

Solutions:

Verify your API key is correct (no extra spaces or characters)
Regenerate your API key if it's expired or compromised:
- Dashboard → Settings → API Keys → Regenerate
Update your MCP client configuration with the new key
Restart your AI client

"Tool Not Found" Error

Solutions:

Check the tool name spelling (e.g., create_strategy not createStrategy)
Verify your MCP client is connected to the server:
- Look for "robonet" in your client's MCP servers list
- Check server status indicator (should be green/connected)
Restart your MCP client to refresh the tool registry

Connection Timeout

Solutions:

Check your internet connection
Verify the MCP server endpoint URL is correct: https://mcp.robonet.finance/mcp?api_key=<API_KEY>
Check if Robonet services are operational (status page or Discord)
Try again in a few minutes (may be temporary network issue)

Strategy Not Showing in Web Interface

Solution: Strategies are linked to your user account via API key. Ensure:

You're logged in with the same account that owns the API key
The strategy was successfully created (check AI response for confirmation)
Refresh the My Strategies page in the web interface

Tips & Best Practices

Efficient Workflow

Start with ideas: Use generate_ideas to explore market-driven concepts
Iterate quickly: Create simple strategies first, then enhance
Always backtest: Never deploy without backtesting
Optimize wisely: Only optimize parameters that impact strategy logic
Enhance selectively: Allora predictions work best for trend-following strategies

Cost Management

Free tools first: Browse strategies, symbols, and indicators without cost
Batch operations: Create multiple strategy variations in one conversation
Reuse backtests: Review existing backtest results before running new ones
Monitor credits: Check your balance regularly in the dashboard

Strategy Development

Descriptive names: Use clear names like RSI_MeanReversion_M (M = medium risk)
Version control: MCP preserves strategy history automatically
Test thoroughly: Backtest across different market conditions (bull, bear, sideways)
Risk management: Always include stop-loss and position sizing logic

Next Steps

Learn more about MCP tools: MCP Tools Reference
Understand backtesting: Backtesting Guide
Deploy your strategy: Deployment Guide
Explore Allora integration: Allora Network Guide

Support

Discord: Join our community
Email: support@robonet.finance

MCP Server Setup ​

Quick Start ​

1. Get Your API Key ​

2. Configure Your Client ​

3. Verify Setup ​

Overview ​

Example Workflow ​

1. Generate Strategy Ideas ​

2. Create a Strategy ​

3. Run a Backtest ​

4. Optimize Parameters ​

5. Enhance with Allora ​

6. Deploy to Production ​

Available MCP Tools ​

Troubleshooting ​

"Insufficient Credits" Error ​

"Invalid API Key" Error ​

"Tool Not Found" Error ​

Connection Timeout ​

Strategy Not Showing in Web Interface ​

Tips & Best Practices ​

Efficient Workflow ​

Cost Management ​

Strategy Development ​

Next Steps ​

Support ​

MCP Server Setup

Quick Start

1. Get Your API Key

2. Configure Your Client

3. Verify Setup

Overview

Example Workflow

1. Generate Strategy Ideas

2. Create a Strategy

3. Run a Backtest

4. Optimize Parameters

5. Enhance with Allora

6. Deploy to Production

Available MCP Tools

Troubleshooting

"Insufficient Credits" Error

"Invalid API Key" Error

"Tool Not Found" Error

Connection Timeout

Strategy Not Showing in Web Interface

Tips & Best Practices

Efficient Workflow

Cost Management

Strategy Development

Next Steps

Support