Enso — An AI sandbox that ships

Every answer is an interactive app. Every sprint makes it smarter. You own the entire factory.

What is Enso?

Enso is an open-source AI platform that doesn't just chat — it builds. Ask a question
and get a structured research dashboard, not a wall of text. Describe an app and watch
Claude Code build it in a live terminal. Run /evolve and an AI team improves your entire
installation autonomously.

Self-hosted. Self-evolving. Yours.

Why Enso?

Get Started

# 1. Clone
git clone https://github.com/Proxy2021/Enso.git && cd Enso

# 2. Setup (installs deps, configures API keys, personalizes your app)
./setup          # macOS / Linux
.\setup.ps1      # Windows PowerShell

Setup handles everything interactively: API keys, Claude Code auth, remote access, app personalization, build, and launch. See SETUP.md for the full guide.

Three Pillars

🏭 Own the Factory

Self-hosted. Open source. Your code, your data, your API keys. No subscriptions. No vendor
lock-in. During setup, Claude Code personalizes the entire app based on who you are — a
developer gets "Forge," an investor gets "Signal," a researcher gets "Nexus."

See PERSONALIZATION-SHOWCASE.md for examples.

📱 Every Answer Is an App

Research questions become interactive dashboards. Build requests become running apps.
File operations become a desktop-grade file manager. 15+ built-in app types, plus custom
apps from a single command.

🧬 AI That Evolves Itself

Type /evolve and an AI team — project leader, architect, engineers, QA — runs a full
improvement sprint. Sprint scores: 4.5 → 7.5 over 4 cycles. No other developer tool does this.

Built-in Capabilities

How It Works

Your message
    │
    ▼
┌─────────────────────────────────────────────────────┐
│  Task Router (auto-classification)                  │
│                                                     │
│  "What's the weather?" ──→ Simple    → Agent chat   │
│  "Fix this bug"        ──→ One-off   → Claude Code  │
│  "Build a CRM system"  ──→ Complex   → Multi-agent  │
│                              orchestration with DAG  │
└──────────────────────────────┬──────────────────────┘
                               │
                               ▼
                    Tool Result + Template
                               │
                               ▼
                  Interactive React App (instant)

No commands. No configuration. The system figures out the right level of effort.

Tech Stack

Frontend: React 19, Zustand 5, Tailwind CSS 4, Vite 6, Capacitor (Android)
Backend: Express, WebSocket, TypeScript, node-pty
AI: Claude Code, Multi-provider LLM (Claude, Gemini, GPT, DeepSeek, Ollama, OpenRouter)
Mobile: Capacitor Android APK + PWA (offline-capable)

Architecture

src/                          React frontend
├── cards/                    Card renderers (DynamicUI, Terminal, Shell, Orchestration, Mission)
├── components/               Timeline, ChatInput, CardContainer, ConnectionPicker, MemoryPanel
├── store/chat.ts             Zustand state (cards, streaming, connections, evolution)
└── lib/                      WS client, JSX sandbox (Sucrase), EnsoUI (17 components), connections

server/              OpenClaw channel plugin (backend)
├── apps/                     Shipped app packages (app.json + template.jsx + executors/)
└── src/
    ├── server.ts             Express + WS server with auth
    ├── channel.ts            OpenClaw ChannelPlugin implementation
    ├── inbound.ts            Message routing (browser → OpenClaw dispatch)
    ├── outbound/             Response delivery, card actions, enhancement, context
    ├── llm-provider.ts       Multi-provider LLM abstraction (callChatLLM)
    ├── task-router.ts        3-tier message classifier (simple/one-off/orchestrated)
    ├── orchestrator.ts       Multi-agent planner (goal → task DAG) with inline targeting
    ├── orchestrator-engine.ts  DAG executor with parallel agents
    ├── card-evolution.ts     Card-type-specific evolution via orchestration
    ├── card-summarizer.ts    Universal content extraction + text/podcast generation
    ├── researcher-tools.ts   Two-phase streaming research pipeline
    ├── build-via-claude.ts   Natural language → app build via Claude Code
    ├── claude-code.ts        Claude Code CLI integration (NDJSON streaming)
    └── tool-factory.ts       Template refinement, auto-heal executor

shared/types.ts               WebSocket protocol types (shared frontend ↔ backend)

Documentation

Contributing

We welcome contributions of all sizes — bug fixes, new app types, UI improvements, docs, translations. See CONTRIBUTING.md for guidelines.

License

MIT License — free forever.