An agent in a box, one command away.
Run Claude Code, Codex or any of 26 coding agents in a Luxor sandbox. One command makes the box, starts the agent and drops you in. Close the laptop and it keeps working. Pick it up again from your phone. Let it run with approvals off, in a box that is nobody's machine.
$curl -fsSL https://luxcode.org/install.sh | sh
Installs lux into ~/.local/bin. Needs the Luxor CLI and jq. llms.txt for agents.
26 coding agents, each in its own sandbox: 21 in the terminal, 5 as web UIs
The box, and the agent in it.
Luxor gives you the sandbox. luxcode turns it into a coding agent you can reach from anywhere.
The box is just there
You never create or resume a sandbox by hand. Run any lux command and the box is simply there: made if it does not exist yet, woken if it was paused, waited for if it is still starting. New boxes start from a golden image with tmux already in it. A persistent volume at ~/work outlives the box, and rm keeps it: name the agent again and it comes back on its files.
Your agent, your models
Pick a terminal agent or a web UI, then where it thinks: the agent's own sign-in, or Nebius, Boundless, Baseten, Friendli and OpenRouter. No config file to edit, and the key never touches your shell history. The agent runs in tmux, so it survives disconnects and pauses; lux <agent> drops you back into the same conversation.
Every coding agent. Each one gets its own box.
lux new asks terminal or web first, then which agent. Each gets a sandbox named after it, independent of the others.
Installed on first use
Claude Code, Codex and opencode come with the base image; the rest install into the box the first time, once.
Luxor's own images
Pi, Droid, Grok Build, Amp, Prime Agent, TinyFish, Anchor and Hotdata ship as Luxor sandbox images. The box starts from the image, pinned and ready, with your key or a sign-in inside.
Private by default
lux opens an ssh tunnel and your browser on it. lux <agent> url makes a public Luxor link, and anyone with it gets in. Terminal agents get the same: a browser terminal on the agent's session, behind a password.
The agent's own sign-in, or bring your own models
The providers in the second half of the table work with every agent that takes a custom endpoint. Keys are asked for once and kept on your machine, mode 0600.
| Provider | Key | Default model | Note |
|---|---|---|---|
| Each agent's default — its own sign-in or key, set once inside the box | |||
| Anthropic | sign in inside the box | Claude's own | Claude Code — the only provider with Remote Control (claude.ai/code, from your phone) |
| OpenAI | codex login inside | Codex's own | Codex |
| opencode | configured inside opencode | opencode's own | opencode and opencode-web; any provider in opencode's catalog |
| GitHub | /login inside | Copilot's own | Copilot — needs a Copilot plan |
| Google sign-in inside | Gemini's own | Gemini CLI | |
| Qwen | Qwen OAuth inside | Qwen's own | Qwen Code |
| Moonshot | kimi login inside | Kimi's own | Kimi Code |
| Factory | FACTORY_API_KEY, or droid login inside | Droid's own | Droid |
| xAI | XAI_API_KEY, or grok login inside | Grok's own | Grok Build |
| Amp | AMP_API_KEY, or amp login inside | Amp's own | Amp |
| Prime Intellect | PRIME_API_KEY, or sign in inside | Prime's own | Prime Agent |
| TinyFish | TINYFISH_API_KEY | — | TinyFish, a web-agent toolkit |
| Anchor Browser | ANCHOR_API_KEY | — | Anchor, a browser-automation toolkit |
| Hotdata | HOTDATA_API_KEY (and HOTDATA_WORKSPACE if your account needs it) | — | Hotdata, a data toolkit |
| The app's own setup | configured inside the app | its own | Crush, Goose, Pi, oh-my-pi, adal, Hermes, OpenClaw, Open WebUI |
| Bring your own models — one key, works with every agent that takes a custom endpoint | |||
| Nebius Token Factory | NEBIUS_API_KEY | moonshotai/Kimi-K3 | no Anthropic API: Claude Code goes through a translation proxy in the box |
| Boundless | BOUNDLESS_API_KEY | claude-fable-5.1 | Anthropic and OpenAI APIs |
| Baseten | BASETEN_API_KEY | moonshotai/Kimi-K3 | Anthropic API in beta |
| Friendli | FRIENDLI_TOKEN | zai-org/GLM-5.3 | |
| OpenRouter | OPENROUTER_API_KEY | anthropic/claude-fable-5.1 | every model behind one key |
Three commands from zero to a running agent.
curl -fsSL https://luxcode.org/install.sh | sh lux into ~/.local/bin lux login your Luxor account, once lux pick an agent, or make one, and you are in
lux pick an agent (or create one) and jump into it lux new [cli|web|harness] create: terminal or web → agent → provider → model → size → volume → name lux <agent> [action] a specific agent by name (or a harness name: your agent for it) up (default) create or resume the box, start the agent, attach shell plain shell in the box run -- <prompt> one-shot headless prompt status agent, provider, box state provider | model switch inference provider, or model size | volume box size (next create) · persistent volume at ~/work url [off] a public link: the web UI, or a browser terminal — anyone with it gets in auth sign in, or store or replace the provider key down | pause pause the box (woken by the next command) rm [--purge] terminate the box; the volume is kept unless --purge lux ls agents, their boxes, and kept volumes lux volumes [rm <name>] every volume; delete one for good lux config [<name> <value>] settings and keys; save a default lux auth [provider] store or replace a provider key lux image [build] the golden image new boxes start from lux pause --all · lux rm --all [--purge]
- small2 vCPU · 4 GiB
- medium4 vCPU · 8 GiB default
- large8 vCPU · 16 GiB
- xlarge12 vCPU · 32 GiB
- xxlarge16 vCPU · 64 GiB
A box is not resized in place: size applies on the next create.
- where
~/work, the agent starts there - sizes10, 25 or 50 GiB
- survivespause,
rm, resizing - kept
rmkeeps it;lux <name>brings the agent back on it - fixeda volume keeps its size for life
Pick generously. lux volumes lists them; lux volumes rm deletes one for good.
lux nebius-codex run -- "fix the failing test"
Runs one prompt and exits, with the provider wired in.
lux tinyfish run -- fetch content get https://…
Toolkits take raw arguments for their own CLI.
Boxes pause by themselves and wake on your next command
Luxor pauses a box two hours after it starts or wakes, even mid-task: that is Luxor's limit, not a setting. Every lux command wakes a paused box, and the clock starts again. tmux and the agent's conversation survive a pause; /tmp and open connections do not. For a job that must run unattended for longer, LUX_STICKY=1 makes a box that never pauses itself, and so never stops costing until you pause it.
Saved with lux config <name> <value>, or set in the environment for one run.
LUX_HARNESS | claude | agent used when you omit one |
LUX_SIZE | medium | default box size offered |
LUX_VOLUME_GB | 10 | default volume size offered (0 = none) |
LUX_DISK_GB | 20 | root disk |
LUX_IMAGE | — | force one registry image for every agent without its own |
LUX_MAX_DURATION | 2h | pause this long after each create or resume (Luxor's cap) |
LUX_PAUSE_RETENTION | 720h | how long a paused box is kept |
LUX_STICKY | 0 | 1 creates boxes that never pause by themselves |
Your Luxor account, your keys, your sandbox.
luxcode is free and open source: one bash script around the Luxor CLI.