f1-nexus-mcp

Model Context Protocol (MCP) server for F1 Nexus - Enable AI agents to optimize F1 race strategies through standardized tools.

Features

• 🤖 AI Agent Integration: Connect Claude, GPT-4, and other LLMs to F1 strategy optimization
• 🔧 MCP Tools: Standardized tools for strategy optimization, simulation, and analysis
• 🌐 Dual Transport: Support for both stdio and Server-Sent Events (SSE)
• 🌦️ Weather API Integration: Real-time weather data from OpenWeatherMap
• 📊 Vector Search: Query historical race data using semantic search
• 🏁 Multi-Agent Consensus: Combine strategies from multiple AI agents

Installation

[dependencies]
f1-nexus-mcp = "1.0.0-alpha.2"

MCP Tools

optimize_strategy

Optimize pit stop strategy for current race conditions.

Input:

{
  "track_id": "monaco",
  "total_laps": 78,
  "current_lap": 15,
  "tire_age": 14,
  "fuel_remaining": 95.5,
  "position": 3,
  "available_compounds": ["C1", "C2", "C3"]
}

Output: Optimal pit stops, tire compounds, predicted race time

simulate_race

Run Monte Carlo simulations for strategy validation.

Input: Strategy + weather + variance parameters Output: Distribution of finish times, DNF probability, confidence intervals

predict_tire_life

ML-based tire degradation prediction.

Input: Current tire data + track conditions Output: Remaining laps, degradation rate, optimal pit window

get_weather_forecast

Get real-time weather data for F1 circuits.

Input: Circuit name or coordinates Output: Temperature, humidity, precipitation, wind, track conditions

query_historical

Semantic search over historical race data.

Input: Natural language query Output: Relevant historical strategies and outcomes

get_agent_consensus

Multi-agent strategy voting and consensus.

Input: Multiple strategy proposals Output: Consensus strategy with confidence scores

Quick Start

Stdio Transport (for Claude Desktop, etc.)

use f1_nexus_mcp::*;

#[tokio::main]
async fn main() -> Result<()> {
    let server = MCPServer::new(Transport::Stdio);
    server.run().await?;
    Ok(())
}

SSE Transport (for web clients)

use f1_nexus_mcp::*;
use axum::{Router, routing::get};

#[tokio::main]
async fn main() -> Result<()> {
    let app = Router::new()
        .route("/mcp/sse", get(sse_handler))
        .route("/mcp/messages", post(message_handler));

    axum::Server::bind(&"0.0.0.0:3000".parse()?)
        .serve(app.into_make_service())
        .await?;

    Ok(())
}

Claude Desktop Integration

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "f1-nexus": {
      "command": "f1-nexus-mcp",
      "args": ["--stdio"],
      "env": {
        "OPENWEATHERMAP_API_KEY": "your-api-key-here"
      }
    }
  }
}

Then ask Claude: "Optimize pit strategy for Monaco GP, starting P3 with 78 laps"

Supported Circuits

Real-time weather data for all 24 F1 circuits including Monaco, Silverstone, Spa, Monza, Suzuka, Singapore, and more.

Use Cases

• AI Race Engineer: Let AI agents make real-time strategy decisions
• Strategy Analysis: Use LLMs to analyze and compare different strategies
• What-if Scenarios: Ask AI to explore edge cases and alternative strategies
• Educational Chatbots: Build F1 strategy learning assistants
• Voice Assistants: "Alexa, optimize my pit strategy for wet conditions"

Documentation

• Full API docs: docs.rs/f1-nexus-mcp
• MCP Specification: modelcontextprotocol.io

Related Crates

• f1-nexus-strategy - Strategy optimization engine
• f1-nexus-core - Core F1 types
• f1-nexus-cli - Command-line interface

License

Licensed under either of Apache License, Version 2.0 or MIT license at your option.