Understanding computers
Section titled “Understanding computers”A Hoody computer is a full Linux computer, running Debian 13 by default. It runs systemd and has its own filesystem and its own network. Every Kit service has its own URL. Your own HTTP server listening on a port is reachable through the http-{port} service label (serve your own app); files and databases are reached through files-1 and sqlite-1. It is closer to a machine than to a Docker image: you can install Docker, Podman or LXC inside it to run your own containers, or, on a dedicated server, run a virtual machine inside it.
The primary interface is HTTP rather than SSH or a deploy pipeline, so you fetch a computer instead of shelling into it. Every service URL is HTTPS over HTTP/1.1 and HTTP/2, and HTTP/3 on servers where it is turned on. The proxy issues and renews the certificates; there is nothing to configure. Deploying to a computer the conventional way still works too.
By the end of this page you have a computer, its service URLs and a snapshot to return to. Set HOODY_TOKEN to your auth token (create one), PROJECT_ID to a project ID (hoody projects list shows yours) and SERVER_ID to a server ID (where it comes from). In the SDK example, define both as strings. After creation, set COMPUTER_ID to the new computer’s ID for the lifecycle examples.
Step 1: Create a computer
Section titled “Step 1: Create a computer”- In the Hub, open Computers and choose Get a computer.
- Under Start from, keep A new computer. Enter a Name and pick a Server and Project.
- Choose Get computer.
# Create a computer in your projecthoody computers create --project "$PROJECT_ID" --name "backend" --server-id "$SERVER_ID"
# List your computershoody computers listimport { HoodyClient } from 'hoody-sdk';const client = new HoodyClient({ baseURL: 'https://api.hoody.com', token: process.env.HOODY_TOKEN });
const computer = await client.api.computers.create(PROJECT_ID, { server_id: SERVER_ID, name: 'backend'});
// running, or still creating: its services start once it is runningconsole.log(computer.data.id);curl -X POST "https://api.hoody.com/api/v1/projects/$PROJECT_ID/computers" \ -H "Authorization: Bearer $HOODY_TOKEN" \ -H "Content-Type: application/json" \ -d '{"server_id": "'"$SERVER_ID"'", "name": "backend"}'The same call as a single GET URL. It runs through the curl-1 service of a
computer you already have, so the first computer comes from one of the other
tabs. Every computer after that can come from a link.
https://PROJECT_ID-OTHER_COMPUTER_ID-curl-1.SERVER.computers.hoody.com/api/v1/curl/request?url=https://api.hoody.com/api/v1/projects/PROJECT_ID/computers&method=POST&bearer_token=HOODY_TOKEN&json={"server_id":"SERVER_ID","name":"backend"}&response=transparentOne request, one link
cURL runs inside your computer and can wrap any HTTP request into a single GET URL. The call stops being something you need a client for and becomes something you can paste into a browser, send in a chat, bookmark, schedule with cron, or drop into a no-code tool.
Nothing is installed on the machine that opens it. The link does carry whatever credentials the call needs, so treat it as you would treat those credentials. It also appears in the cURL service's request log, not only the target's. Prefer a delegated token with minimal permissions and an expiry: see API tokens.
Slashes, colons and braces pass through as they are. The one character you must
encode is an & inside a value, which happens when the wrapped URL
carries its own query string. Left raw it ends the value early, and the rest is
read as cURL's own parameters, so you get a 200 on a request you did
not make.
How the wrapping works Chaining calls into one link Turning a link into a shortcut
The response is the computer, running or still creating. Its Hoody Kit services start once it is running, so give them a short wait. For this walkthrough, your auth token must have computers.features.hoody_kit and computers.features.ai. With an auth token, an omitted hoody_kit or ai is turned on only when the token has the matching permission; without it, the computer is created with that feature off. Sending hoody_kit: true or ai: true without the permission gets 403 (what a computer includes).
Step 2: Open its service URLs
Section titled “Step 2: Open its service URLs”Every service in every computer answers on the same hostname shape, and only the service label changes between them:
https://{projectId}-{computerId}-{service}[-{instance}].{server}.computers.hoody.comSubstituting a real project, computer and server gives the shell for that computer. Both IDs are 24 hexadecimal characters:
https://67e89abc123def456789abcd-890abcdef12345678901cdef-terminal-1.node-us.computers.hoody.comSwap terminal-1 for another label: files-1, sqlite-1, display-1, browser-1, exec-1, agent-1, code-1, curl-1, daemon-1, cron-1, n-1, pipe-1, notes-1, watch-1, run-1, tunnel-1, logs-1, egress or vm-1. What each one does: Kit services. hoody computers urls list prints every Kit URL of a computer. In the Hub, open Computers, select the computer and open its Connect tab: Program links lists each service’s URL with buttons to open or copy it.
The URLs open from any device with a browser, and the Hoody CLI drives them from any terminal. ssh hoody.com opens the hosted Hoody CLI. After you sign in, an account with a computer opens the Hub; an account without one first offers a free computer. Before you share a URL, read who can open it: open by default.
A computer can run several instances of the same service. Three terminals, two SQLite URLs, and a browser:
https://PROJECT_ID-COMPUTER_ID-terminal-1.node-us.computers.hoody.comhttps://PROJECT_ID-COMPUTER_ID-terminal-2.node-us.computers.hoody.comhttps://PROJECT_ID-COMPUTER_ID-terminal-3.node-us.computers.hoody.comhttps://PROJECT_ID-COMPUTER_ID-sqlite-1.node-us.computers.hoody.comhttps://PROJECT_ID-COMPUTER_ID-sqlite-2.node-us.computers.hoody.comhttps://PROJECT_ID-COMPUTER_ID-browser-1.node-us.computers.hoody.comAll six run in the same computer. Each terminal instance has its own URL, its own process and its own state. The two SQLite URLs reach the same SQLite service; use ?db=app and ?db=scratch to select different databases.
Step 3: Start, stop and snapshot it
Section titled “Step 3: Start, stop and snapshot it”Snapshot before you make changes, and restore if something breaks. A restore discards the current filesystem and puts the snapshot’s in its place, so the computer’s current state is lost; a running computer restarts. Restore takes the snapshot’s canonical name, not its alias (naming). Here the name is before-experiment: set SNAPSHOT_NAME to it before restoring (in the SDK, define it as a string). To look names up later, run hoody snapshots list -c "$COMPUTER_ID"; the HTTP and SDK create responses carry it at data.snapshot.name. Before your first restore, read Restore a snapshot.
- In the Hub, open Computers and select the computer. Choose Start or Stop in its menu.
- To snapshot it, open Backups and choose Take snapshot. Enter
before-experimentas the Name, then choose Take snapshot. - To restore, choose Restore on the snapshot’s row, then Restore computer.
# Start a stopped computerhoody computers start "$COMPUTER_ID"
# Stop a running computerhoody computers stop "$COMPUTER_ID"
# Snapshot before making changeshoody snapshots create -c "$COMPUTER_ID" --alias "before-experiment"
# Restore if something breaks (the CLI asks you to confirm; add -y to skip the prompt)hoody snapshots restore "$SNAPSHOT_NAME" -c "$COMPUTER_ID"// Startawait client.api.computers.start(COMPUTER_ID);
// Stopawait client.api.computers.stop(COMPUTER_ID);
// Snapshotconst snapshot = await client.api.snapshots.create(COMPUTER_ID, { alias: 'before-experiment'});
// Restoreawait client.api.snapshots.restore(COMPUTER_ID, SNAPSHOT_NAME);# Startcurl -X POST "https://api.hoody.com/api/v1/computers/$COMPUTER_ID/start" \ -H "Authorization: Bearer $HOODY_TOKEN"
# Stopcurl -X POST "https://api.hoody.com/api/v1/computers/$COMPUTER_ID/stop" \ -H "Authorization: Bearer $HOODY_TOKEN"
# Snapshotcurl -X POST "https://api.hoody.com/api/v1/computers/$COMPUTER_ID/snapshots" \ -H "Authorization: Bearer $HOODY_TOKEN" \ -H "Content-Type: application/json" \ -d '{"alias": "before-experiment"}'
# Restorecurl -X PUT "https://api.hoody.com/api/v1/computers/$COMPUTER_ID/snapshots/$SNAPSHOT_NAME" \ -H "Authorization: Bearer $HOODY_TOKEN"Each action as a single GET URL. Route them through a different running
computer’s curl-1 (OTHER_COMPUTER_ID below): a stopped computer cannot
serve the request that starts it.
# Starthttps://PROJECT_ID-OTHER_COMPUTER_ID-curl-1.SERVER.computers.hoody.com/api/v1/curl/request?url=https://api.hoody.com/api/v1/computers/COMPUTER_ID/start&method=POST&bearer_token=HOODY_TOKEN&response=transparent
# Stophttps://PROJECT_ID-OTHER_COMPUTER_ID-curl-1.SERVER.computers.hoody.com/api/v1/curl/request?url=https://api.hoody.com/api/v1/computers/COMPUTER_ID/stop&method=POST&bearer_token=HOODY_TOKEN&response=transparent
# Snapshothttps://PROJECT_ID-OTHER_COMPUTER_ID-curl-1.SERVER.computers.hoody.com/api/v1/curl/request?url=https://api.hoody.com/api/v1/computers/COMPUTER_ID/snapshots&method=POST&bearer_token=HOODY_TOKEN&json={"alias":"before-experiment"}&response=transparent
# Restorehttps://PROJECT_ID-OTHER_COMPUTER_ID-curl-1.SERVER.computers.hoody.com/api/v1/curl/request?url=https://api.hoody.com/api/v1/computers/COMPUTER_ID/snapshots/SNAPSHOT_NAME&method=PUT&bearer_token=HOODY_TOKEN&response=transparentOne request, one link
cURL runs inside your computer and can wrap any HTTP request into a single GET URL. The call stops being something you need a client for and becomes something you can paste into a browser, send in a chat, bookmark, schedule with cron, or drop into a no-code tool.
Nothing is installed on the machine that opens it. The link does carry whatever credentials the call needs, so treat it as you would treat those credentials. It also appears in the cURL service's request log, not only the target's. Prefer a delegated token with minimal permissions and an expiry: see API tokens.
Slashes, colons and braces pass through as they are. The one character you must
encode is an & inside a value, which happens when the wrapped URL
carries its own query string. Left raw it ends the value early, and the rest is
read as cURL's own parameters, so you get a 200 on a request you did
not make.
How the wrapping works Chaining calls into one link Turning a link into a shortcut
How does a computer differ from a Docker container?
Section titled “How does a computer differ from a Docker container?”| Feature | Docker | Hoody computers |
|---|---|---|
| Base system | Minimal layers | Debian 13 with systemd (images) |
| Networking | Internal bridge, port mapping | Every service has a public URL |
| Access method | docker exec / SSH | HTTP from anywhere |
| Built-in services | None; you bring everything | The Hoody Kit HTTP services |
| Collaboration | Not designed for it | Multiplayer by default |
| Snapshots | Images by docker commit, without volume data | Full filesystem snapshots and restore |
| Multiple instances | Separate containers | terminal-1, terminal-2… in the same computer |
Docker containers are build artifacts; Hoody computers are full computers.
What does a computer cost?
Section titled “What does a computer cost?”Pricing is per server, not per computer: each plan is one server, and the computers on it are not charged (Pricing). Computers have no per-unit cost. Once a server exists, you can create as many as its capacity and limits allow, including your project’s computer limit (max_computers).
Under per-VPS pricing, three environments at $40 a month each cost $120 a month. On Hoody, one server carries all of them. That removes the usual reason to ration environments: a dev computer, a staging copy, an experiment and an AI playground can each get their own computer.
Advanced
Section titled “Advanced”- Restart, pause and force-stop: the other lifecycle operations, and starting computers when their server boots, are on Managing computers.
- Change or delete a computer: an update changes its name, environment variables or SSH key, and a delete removes it (Create, Edit, Delete).
- Copy a computer: duplicate one into another project or onto another server, for staging or an isolated experiment (Copy and Sync).
- A plain computer without the Kit:
hoody_kit: falsegives Linux without the Kit’s HTTP services: you manage it over SSH with a key, and your own HTTP servers are still reachable throughhttp-{port}(what a computer includes). - Virtual machines inside a computer:
vm-1runs QEMU virtual machines and needs/dev/kvm, which KVM passthrough gives a computer on a dedicated server only (Virtual Machines).
Reference
Section titled “Reference”Every Kit service a computer includes, with its URL label and what it does: Understanding computers reference.
What’s next
Section titled “What’s next”- The Hoody Proxy: who can open these URLs
- Managing computers: every lifecycle operation
- Computers: what each lifecycle state means
- The Hoody Kit: each service in depth