Your AI-assisted workspace for everything.

Projects, sessions, context packs, secrets, clusters, and streaming AI chat — all in one place.

Orbit UI demo showing the explorer, workspace, and AI chat

Everything you need, one workspace.

Orbit brings AI, context, and infrastructure together so you can focus on building.

Streaming AI Chat

Real-time streaming responses powered by Claude via Vertex AI or Anthropic API. Context-aware conversations that understand your project.

Context Packs

Bundle related context together and pin it to sessions. The AI receives structured context on every message for accurate assistance.

Sessions & Workspaces

Organize work into projects and sessions. Each session maintains its own context layers, making it easy to switch between tasks.

Secrets Management

Store and manage API keys, credentials, and sensitive configuration securely. Access them across sessions without exposure.

MCP Integration

Connect external tools via the Model Context Protocol. Bring in data from Jira, GitHub, and more directly into your AI conversations.

Cluster Management

Manage OpenShift clusters from your workspace. Deploy, monitor, and scale your applications with the orbit-operator.

Your AI companion, always by your side.

Orbi reacts to what you're doing — thinking when the AI streams, happy when it finishes, and even sleeps when you step away.

Zzz
Idle
Thinking
Happy
Zzz
Sleeping

Orbi dog visual based on CSS Husky by Harshal Parmar

Built on proven foundations.

A modern stack designed for performance, scalability, and developer experience.

Frontend
React + TypeScript
Vite · Tailwind CSS
Backend
Python · FastAPI
SQLAlchemy · Alembic
Data
PostgreSQL · pgvector
Redis · Celery
AI
Claude (Anthropic)
Vertex AI · MCP
Deploy
OpenShift · Podman
Docker Compose

Up and running in minutes.

Clone the repo and start building with AI.

1

Clone the repo

Grab the source code and enter the directory.

git clone https://github.com/GowthamShanmugam/orbit.git && cd orbit
2

Start services

Spin up Postgres and Redis with one command.

podman compose up -d postgres redis
3

Configure

Copy the example env and set your AI provider key.

cp .env.example .env
4

Launch

Start the backend and frontend, then open the app.

http://localhost:5173

Ready to orbit?

Open source. Self-hosted. Built for developers who want AI that understands context.