DevEx MCP Server

DevEx MCP Server

AI-powered cloud development IDE, so you can run AI to surf in a secure sandbox environment.

Category
Visit Server

README

<h1 align="center">⚡ DevEx – The Open Cloud IDE Platform</h1>

<p align="center"> <b>Run full-featured, containerized dev environments in the cloud — with AI assistance and secure sandboxing.</b><br/> Scalable REPL sessions powered by Kubernetes, WebSockets, and GoLang. </p>

<p align="center"> <a href="https://github.com/ParthKapoor-dev/devex/stargazers"> <img src="https://img.shields.io/github/stars/ParthKapoor-dev/devex?style=for-the-badge" /> </a> <a href="https://github.com/ParthKapoor-dev/devex/issues"> <img src="https://img.shields.io/github/issues/ParthKapoor-dev/devex?style=for-the-badge" /> </a> <a href="https://github.com/ParthKapoor-dev/devex/blob/main/LICENSE"> <img src="https://img.shields.io/github/license/ParthKapoor-dev/devex?style=for-the-badge" /> </a> </p>

<p align="center"> <img src="https://raw.githubusercontent.com/parthkapoor-dev/devex/main/assets/devx.png" alt="Cloud Dev IDE Banner" /> </p>


DevEx is an open-source platform to spin up cloud-based development environments (REPLs) that work like your local setup — but in the browser, anywhere, anytime.

Built for flexibility, you can code in your preferred stack, connect over SSH, or run sessions with AI agents via our MCP server. We also support sandboxing AI-generated code securely, letting you test real changes in isolated environments.

Unlike Gitpod or E2B, DevEx is lightweight, self-hostable, and production-grade out of the box — with a developer-first CLI, seamless AI integration, and Kubernetes-native design for infinite scale.

📣 Want to add your own REPL template? It's easy! 👉 Read the Contributing Guide to get started.


<p align="center"> <a href="https://glama.ai/mcp/servers/@ParthKapoor-dev/devex"> <img width="380" height="200" src="https://glama.ai/mcp/servers/@ParthKapoor-dev/devex/badge" /> </a> </p>

<p align="center"><b>🚀 MCP Server is now live at <a href="https://glama.ai">glama.ai</a> – enabling AI agents to interact directly with your REPLs!</b></p>

✨ Features

  • 🔐 GitHub OAuth authentication
  • 🪄 Create, Start, Stop, Delete REPLs via Core API
  • 💾 S3-backed file persistence
  • 📦 Kubernetes Deployments per REPL (Dynamic)
  • 📡 WebSocket-based Editor & Terminal
  • 🧹 Ephemeral containers for cleanup and sync
  • 🔒 HTTPS & TLS via Let's Encrypt & Cert Manager
  • 🎨 Beautiful Next.js + Tailwind frontend

⚙️ How It Works

graph TB
    User[👤 User] --> Web[🌐 Web Frontend<br/>React/Next.js Application]
    Web --> Core[🔧 Core Backend<br/>• User Authentication<br/>• Repl Management<br/>• S3 Integration<br/>• K8s Orchestration]
    Core --> S3[(🗄️ S3 Storage<br/>username/repl-id/<br/>├── templates/<br/>└── user-files/)]

    subgraph K8sCluster["☸️ Kubernetes Cluster"]
        direction TB
        IngressController[🚪 Ingress NGINX Controller<br/>Traffic Routing]
        CertManager[🔒 Cert Manager<br/>TLS Certificate Management]

        subgraph ReplResources["📦 Per-Repl Resources"]
            Deployment[🚀 Deployment<br/>Repl Container Instance]
            Service[🔗 Service<br/>Internal Network Access]
            Ingress[🌍 Ingress<br/>External Access Route]
        end

        subgraph Pod["🏠 Repl Pod"]
            MainContainer[🐳 Runner Container<br/>• WebSocket Server<br/>• File Operations<br/>• PTY/Terminal Access<br/>• Code Execution]
            EphemeralContainer[⚡ Ephemeral Container<br/>File Sync Back to S3<br/>🔄 Cleanup Process]
        end

        Deployment --> Pod
        Service --> Pod
        Ingress --> Service
        IngressController --> Ingress
    end

    Core --> K8sCluster
    Core -.->|Create Resources<br/>Deploy → Service → Ingress| ReplResources
    Web -.->|🔌 WebSocket Connection<br/>• File Management<br/>• Terminal Access<br/>• Real-time Collaboration| MainContainer
    Core -.->|📁 Copy Template<br/>to user directory| S3
    EphemeralContainer -.->|💾 Sync Files Back<br/>Before Cleanup| S3
    MainContainer -.->|📂 Load Files<br/>on Session Start| S3

🌀 Session Lifecycle

sequenceDiagram
    participant U as User
    participant W as Web Frontend
    participant C as Core Service
    participant K as Kubernetes
    participant S as S3 Storage
    participant R as Runner Container

    U->>W: Create New Repl
    W->>C: POST /repl/create
    C->>S: Create user directory<br/>Copy template files

    U->>W: Start Session
    W->>C: POST /repl/start
    C->>K: Create Deployment<br/>Service & Ingress
    K->>R: Initialize Container
    R->>S: Download files
    R->>W: WebSocket Connection

    Note over U,R: Development Session Active
    U->>R: File operations via WebSocket
    R->>R: Real-time file editing

    U->>W: Close Session
    W->>C: POST /repl/stop
    C->>K: Inject Ephemeral Container
    K->>S: Upload modified files
    C->>K: Delete Resources

🔩 Key Components

web/Frontend

  • Built with Next.js + Tailwind CSS
  • GitHub OAuth login
  • GUI for File Tree, Editor, Terminal
  • WebSocket hooks to interact with Runner

core/Backend API

  • Written in Go
  • Handles user auth, S3 ops, Kubernetes deployments, cleanup
  • Redis for REPL session state
  • 📄 See core/README.md for detailed architecture & deployment steps

runner/REPL Runtime Container

  • Lightweight Go server

  • WebSocket API for:

    • File tree and file content access
    • Terminal (PTY) sessions
  • 📄 See runner/README.md for event list and package internals

k8s/Kubernetes Bootstrap & TLS

  • Contains:

    • Ingress-NGINX setup
    • cert-manager + Let’s Encrypt for auto TLS
  • 📄 See k8s/README.md for full setup instructions

templates/

  • Base folders (e.g. Node.js, Python) copied on REPL creation
  • Language-specific dockerized scaffolds
  • 📦 Want to add your own template? See the Contribution Guide

🧱 Infrastructure

  • ☸️ Kubernetes cluster for REPL pods
  • 🔒 Cert Manager + Let’s Encrypt for TLS
  • 🗃️ S3-compatible storage for persistence
  • 🐳 Docker images for runtime environments
  • ⚙️ Redis for in-memory session tracking

📦 Deployment Flow

  1. User logs in and creates a REPL

  2. core/ copies a template into username/repl-id/ on S3

  3. core/ deploys a pod, service, ingress in Kubernetes

  4. runner/ connects via WebSocket and serves FS + Terminal

  5. On session end:

    • Ephemeral container uploads updated files to S3
    • All K8s resources are cleaned up

💻 Tech Stack

Layer Stack
Frontend Next.js, Tailwind, WebSockets
Backend GoLang (Echo/Fiber), Redis, S3 SDK
Runner GoLang + PTY + WebSocket
Orchestration Kubernetes, Docker, Docker Swarm
Networking Ingress NGINX + cert-manager
Auth GitHub OAuth

📄 Per-Component Docs

📚 For deeper implementation details:


🧠 Why I Built This

“This project is my deep dive into Cloud Infrastructure, DevOps, and FullStack Engineering — wrapped in a real-world application.” — Parth Kapoor


🌐 Demo & Links


🤝 Contributing

This project is under active development. Want to contribute a template, fix, or feature? Start here 👉 CONTRIBUTING.md

Pull requests, suggestions, and feedback are always welcome!


📜 License

Licensed under the MIT License

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured