What is Orbit?
Orbit is an AI-native development environment. A native desktop app where one AI agent works across every surface you build with — editor, browser, terminal, and docs — building its own context from all of them.
The problem
Every AI coding tool connects an agent to one surface. Cursor connects it to a code editor. Claude Code connects it to a terminal. Codex runs tasks asynchronously. The agent writes code — but it can't see the result.
You copy error logs from the browser and paste them into the chat. You screenshot UI bugs and describe them in plain text. You switch between your editor, terminal, browser, and documentation — and the agent only knows what you tell it.
Wrong context means wrong code. The problem isn't code generation speed — it's that the agent can't see what you see.
Orbit's approach
Give the agent access to every surface you build with, and let it build its own context. The agent takes screenshots of your running app. It reads your terminal output. It reads your docs in the vault. It sees your code. You work in one window. The agent works across all of it.
This is the same philosophy behind Claude Code's success — letting the agent use tools to build its own understanding rather than relying on human-curated context. Orbit extends that approach from the terminal to a full visual environment.
What's inside
Every surface is connected to the same agent.
Full conversational AI agent powered by the Claude SDK. Sub-agents, skills, plugins, and a built-in MCP creator. Configurable: plan-first or direct execution.
CodeMirror 6 with full LSP support. Syntax highlighting, code intelligence, file system access. Not VS Code — no extension ecosystem, no inline completions.
Embedded browser inside the environment. The agent takes screenshots of your running app, navigates, clicks, scrolls, and fills forms autonomously. Click any element, describe the change.
Tabbed terminal with full agent access. Run commands, scripts, and manage processes.
Markdown editor for docs, PRDs, notes, and planning. Write your requirements — the agent reads them as context while building.
Staging, commits, branches, merges, diffs, history, push and pull. All from within the environment.
What Orbit is not
Not a code editor with AI bolted on. That's Cursor's starting point — a VS Code fork with an agent panel. Orbit is built from scratch with Tauri and React.
Not a terminal agent. That's Claude Code. Great tool, different model. Orbit gives the agent visual surfaces too.
Not a browser-based builder. That's Bolt, Lovable, Replit. Your project lives on their servers. In Orbit, your project lives on your machine as real files.
Not a VS Code fork. No VS Code extension support. The editor is CodeMirror 6. Orbit has its own skills and plugin system instead.
What to know
No inline completions. No tab-autocomplete like Copilot or Cursor. All AI interaction is through the agent conversation. Different model for different workflows.
Claude-only for V1. Multi-model support (OpenAI, Google, local models) is on the roadmap.
macOS only. Apple Silicon. Windows and Linux are coming.
No deploy button. Orbit builds your app but doesn't deploy it. You'll use Vercel, Railway, or similar. One-click deploy is on the roadmap.
Free during early access. Sign in with your Claude account (even the free plan) or bring your own Anthropic API key. No token markup.
Who it's for
One agent across every surface. Extensible. The agent sees your running app.
For FoundersFrom idea to running app. Write requirements, the agent builds while you direct.
For Product ManagersFrom PRD to working prototype. Direct the agent instead of waiting for engineering.
For Vibe CodersSee your app. Click anything. Describe the change. Done.
Free during early access
Sign in with your Claude account or bring your own API key. No setup required.