Orbi Cloud

Orbi Cloud is the hosted version of Orbi’s software dark factory: you define the goal and acceptance criteria, and Orbi takes a GitHub Issue through implementation, tests, independent review, fixes, and merge. Orbi operates the infrastructure. After you connect a repository, Orbi reads it and suggests work it can do. Review a new task’s draft before handing it off, or dispatch an existing Issue with Let Orbi do this. You can also Write a request yourself or label an Issue ai-ready in GitHub. Orbi delivers the work in an isolated workspace and merges the reviewed change after its merge checks pass. Issues, pull requests and release evidence stay in your repository. Orbi Cloud runs the same delivery engine as self-hosted Orbi, which is open source. The difference is who operates the runner: with Cloud, Orbi does.

​

The whole path, end to end

Authorize and connect a repository, choose the work, and decide when to release. Preparing the environment and reading for suggestions start automatically. Once you hand off a task, Orbi takes it through development and merge. You decide which task starts and when to release. The handoff buttons apply ai-ready for you; reading the repository or opening a suggested draft does not dispatch work. The status page guides you through five setup steps, ending when you hand over the first request. Releasing is optional, and subscribing is not required for the first three merged deliveries. You can also try the official example without using those free deliveries.

​

What you actually do

​

The delivery loop

The review is a separate session that did not write the code. It repairs the findings it can and re-runs the tests; only a clean verdict reaches the merge gate. When automatic recovery is not safe, the Issue lands on ai-blocked with a comment saying why — that state is a deliberate stop, waiting for a human decision, not a crash. Read the delivery lifecycle for what each label means and what moves an Issue between them.

​

What it costs

Your first 3 merged deliveries are free — no card and no subscription. Failed deliveries do not use up that allowance. After that there are two paid plans, Solo and Pro. Both include model usage in the price: Solo with 400,000,000 tokens a month and one repository, Pro with 1,200,000,000 tokens a month and up to five repositories. There is no per-token overage billing: when the month’s allowance runs out, new deliveries pause until the next month, or until you add your own model key. The current prices, the yearly option and any founding offer are on the pricing page. Billing covers how subscribing works, and limits and quotas covers how the allowance is counted.

​

What Orbi Cloud does not do

It does not discover repositories on its own — it only ever works in the repositories you connect. A release follows your version decision and release policy; it runs through an Issue carrying ai-ready and ai-release. Ordinary code changes reach the base through reviewed PRs; release version updates and empty-repository initialization have the documented direct-write paths, and it does not touch a delivery that is already running when the platform is upgraded. Read security for the full boundary list, including where your model key is stored and what the runner can reach. Use Settings in your avatar menu to configure Telegram notifications and Orbi’s message language, independently of the website language.

​

Where to go next

If you want the shortest path to a merged pull request, start with the quickstart. If something is already wrong, go to troubleshooting.

​

The engine’s own documentation

These pages cover Orbi Cloud — the hosted product. The delivery engine underneath is open source and documented separately at docs.orbi.build. Go there when you want the mechanism rather than the product: