Repolode
Submit
casosorg

casosorg/casos

A single-binary cloud OS built on embedded Kubernetes, with a web UI, app store, private AI model hosting and AI agent access.

Open-source alternative to:CoolifyCasaOS

View on GitHubWebsite
Stars
1,383
Forks
104
Open issues
4
Last commit
17 hours ago

What is it?

CasOS runs a real Kubernetes cluster from one executable. The API server, controller manager and scheduler are embedded in the binary, and the machine it starts on becomes a worker node. On Windows that node runs inside WSL, which the installer sets up when it is missing. The project includes a web UI, an app store fed by sealos templates and Helm charts, and managed databases with backups.

Beyond app hosting, CasOS can install Ollama and Open WebUI on a machine with an NVIDIA GPU and serve models through an OpenAI-compatible API at /v1. A built-in MCP server lets tools such as Claude Code, Cursor and Codex deploy apps, read logs and roll back releases. Users can also deploy from a public Git repository, which CasOS builds in the cluster, or paste a docker-compose.yml file to run it.

Multiple computers can join one cloud through an invite, including machines behind NAT, using an embedded Headscale coordination server and Tailscale clients. Sign-in uses a built-in admin account by default, or Casdoor single sign-on when configured.

Key features

  • 01Embedded Kubernetes control plane in a single binary, with no separate cluster needed.
  • 02Automatic worker node setup on Linux and on Windows through WSL2.
  • 03App store for installing Helm releases from the web UI.
  • 04Deploy from a Git repository or a docker-compose.yml file.
  • 05One-click Ollama and Open WebUI with an OpenAI-compatible API.
  • 06MCP server for AI agents, plus browser-based VS Code dev boxes.

Use cases

  • Running a personal or home-lab Kubernetes cluster without writing YAML.
  • Hosting open-source LLMs privately on a local NVIDIA GPU.
  • Letting AI coding agents deploy and inspect apps through MCP.
  • Combining several home or office computers into one cluster over the internet.

Quick start

  1. 1

    Install with the script (Linux/macOS)

    curl -fsSL https://raw.githubusercontent.com/casosorg/casos/master/scripts/install.sh | bash
  2. 2

    Start CasOS

    casos
  3. 3

    Open http://localhost:20080 and change the default admin password

Things to know

  • On macOS, CasOS cannot be a worker node, so you need a Linux machine or VM.
  • A fresh install uses the default admin password 123, so change it right away.
  • On Linux, CasOS must run as root or as a user with passwordless sudo.

Topics

#kubernetes#self-hosted#ai-agent#llm#mcp#golang#react#paas#auth#authorization

FAQ

What is casos?
A single-binary cloud OS built on embedded Kubernetes, with a web UI, app store, private AI model hosting and AI agent access. CasOS runs a real Kubernetes cluster from one executable. The API server, controller manager and scheduler are embedded in the binary, and the machine it starts on becomes a worker node. On Windows that node runs inside WSL, which the installer sets up when it is missing. The project includes a web UI, an app store fed by sealos templates and Helm charts, and managed databases with backups.
Is casos free to use? What license does it have?
casos is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is casos still maintained?
Actively maintained. The last commit to casos was on Oct 19, 2026.
What can I use casos for?
Running a personal or home-lab Kubernetes cluster without writing YAML. Hosting open-source LLMs privately on a local NVIDIA GPU. Letting AI coding agents deploy and inspect apps through MCP. Combining several home or office computers into one cluster over the internet.
How do I install casos?
Install with the script (Linux/macOS): curl -fsSL https://raw.githubusercontent.com/casosorg/casos/master/scripts/install.sh | bash Start CasOS: casos Open http://localhost:20080 and change the default admin password
Where can I get casos?
The source code and installation instructions are on GitHub: https://github.com/casosorg/casos

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects