Fire in da houseTop Tip:Paying $100+ per month for Perplexity, MidJourney, Runway, ChatGPT and other tools is crazy - get all your AI tools in one site starting at $15 per month with Galaxy AI Fire in da houseCheck it out free

ESP32MCPServer

MCP.Pizza Chef: navado

ESP32MCPServer is an MCP server implementation for ESP32 microcontrollers, providing a WebSocket-based interface for real-time resource discovery, monitoring, and WiFi configuration. It supports thread-safe request handling, AsyncWebServer integration, and LittleFS for configuration storage, enabling AI models to connect and interact with ESP32 hardware interfaces efficiently.

Use This MCP server To

Enable AI models to discover and monitor ESP32 resources in real time Configure ESP32 WiFi settings via a web interface Provide thread-safe handling of MCP requests on ESP32 devices Store and retrieve configuration data using LittleFS on ESP32 Integrate ESP32 hardware with AI workflows through WebSocket MCP protocol Test and validate ESP32 MCP server functionality with comprehensive test suite

README

ESP32 MCP Server

A Model Context Protocol (MCP) implementation for ESP32 microcontrollers, providing a WebSocket-based interface for resource discovery and monitoring.

Status: Not Compiling, initial commit as is from the model

Created with Claude 3.5 Sonet on the commit date (with minor obvioud fixes with automatic formating, etc.)

architecture

Features

  • MCP protocol implementation (v0.1.0)
  • WebSocket server for real-time updates
  • Resource discovery and monitoring
  • WiFi configuration via web interface
  • Thread-safe request handling
  • Comprehensive test suite
  • AsyncWebServer integration
  • LittleFS support for configuration storage

Prerequisites

Hardware

  • ESP32 S3 DevKitC-1 board
  • USB cable for programming

Software

  • PlatformIO Core (CLI) or PlatformIO IDE
  • Python 3.7 or higher
  • Git

Architecture

flowchart TD
    Start[Start Application] --> Setup[Setup]
    Setup -->|Initialize Filesystem| InitFS[Initialize LittleFS]
    Setup -->|Start Network| StartNetwork[Initialize Network Manager]
    Setup -->|Create Tasks| CreateTasks[Create and Assign Tasks]

    subgraph Network
        StartNetwork --> APCheck[Check AP or Connect Mode]
        APCheck -->|Credentials Exist| Connect[Connect to WiFi]
        APCheck -->|No Credentials| StartAP[Start Access Point]
        Connect --> NetworkReady[Network Ready]
        StartAP --> NetworkReady
    end

    subgraph MCP_Server
        MCP[Start MCP Server] --> HandleClient[Handle Client Connections]
        HandleClient --> HandleRequest[Handle Requests]
        HandleRequest -->|WebSocket Events| WebSocket[Handle WebSocket]
        HandleRequest -->|HTTP Endpoints| HTTP[Process HTTP Requests]
    end

    subgraph Metrics
        self[Start Metrics System] --> InitMetrics[Initialize System Metrics]
        InitMetrics --> CollectMetrics[Collect Metrics Periodically]
        CollectMetrics --> SaveMetrics[Save Metrics to Filesystem]
    end

    subgraph Logger
        self[Start uLogger] --> LogMetrics[Log Metrics Data]
        LogMetrics --> CompactLogs[Compact Logs if Necessary]
        CompactLogs -->|Rotates Logs| LogRotation
    end

    CreateTasks -->|Network Task| NetworkTask[Run Network Task on Core]
    CreateTasks -->|MCP Task| MCPTask[Run MCP Server Task on Core]
    NetworkTask --> Network
    MCPTask --> MCP_Server
    MCP_Server --> Metrics
    Metrics --> Logger


Loading

Installation

  1. Clone the repository:
git clone https://github.com/yourusername/esp32-mcp-server.git
cd esp32-mcp-server
  1. Install dependencies:
pio pkg install
  1. Build and upload the filesystem:
pio run -t uploadfs
  1. Build and upload the firmware:
pio run -t upload

Usage

Initial Setup

  1. Power on the ESP32. It will create an access point named "ESP32_XXXXXX"
  2. Connect to the access point
  3. Navigate to http://192.168.4.1
  4. Configure your WiFi credentials
  5. The device will connect to your network

MCP Connection

Connect to the MCP server using WebSocket on port 9000:

const ws = new WebSocket('ws://YOUR_ESP32_IP:9000');

// Initialize connection
ws.send(JSON.stringify({
    jsonrpc: "2.0",
    method: "initialize",
    id: 1
}));

// List available resources
ws.send(JSON.stringify({
    jsonrpc: "2.0",
    method: "resources/list",
    id: 2
}));

Testing

Run the test suite:

# Run all tests
pio test -e native

# Run specific test
pio test -e native -f test_request_queue

# Run with coverage
pio test -e native --coverage

Contributing

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/amazing-feature)
  3. Commit your changes (git commit -m 'Add amazing feature')
  4. Push to the branch (git push origin feature/amazing-feature)
  5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

ESP32MCPServer FAQ

How do I set up the ESP32MCPServer on my device?
You need an ESP32 S3 DevKitC-1 board, PlatformIO (CLI or IDE), Python 3.7+, and Git; then follow the build instructions in the repository.
What communication protocol does ESP32MCPServer use?
It uses WebSocket for real-time updates and MCP protocol v0.1.0 for structured interaction.
Can ESP32MCPServer handle concurrent requests safely?
Yes, it implements thread-safe request handling to manage multiple simultaneous connections.
How is configuration data stored on the ESP32?
Configuration is stored using LittleFS, a lightweight filesystem supported by the server.
Does ESP32MCPServer support WiFi configuration?
Yes, it provides a web interface to configure WiFi settings on the ESP32 device.
Is there a test suite available for ESP32MCPServer?
Yes, a comprehensive test suite is included to validate server functionality.
Which development environments are recommended?
PlatformIO Core or PlatformIO IDE are recommended for building and deploying the server.
Can this server be integrated with AI models from providers like OpenAI, Claude, or Gemini?
Yes, it is designed to expose ESP32 interfaces to AI models via MCP, compatible with providers like OpenAI, Claude, and Gemini.