close

The lights-out software factory for the Agent era

File an Issue. Get a release.

Orbi writes code for your GitHub Issues with AI. It arranges independent AI review, merges approved changes, and follows your repository’s release process. You don’t have to supervise every step.

Try Orbi Cloud free →

GitHub sign-in and App install required. No credit card required Open source · Self-host ↗

orbi-build/orbi · real delivery

Issue #1306 → Release v0.5.42

  1. Issue openedstart
  2. PR #1309 reviewed & merged+1h 27m
  3. Released v0.5.42+6h 42m
Same day · every commit written by Orbi
857PRs merged by Orbi on its own repo
143releases shipped
Open sourceAGPL-3.0, self-host free

How Orbi automates
GitHub Issue to tagged release

An Issue is the executable unit, but Orbi does not manage Issues in isolation. Epics and milestones hold scope. Dependencies order the work. Review and repair converge each PR. A Release Issue freezes the result into a version.

01Label an Issue ai-ready
Issueai-readyWorking copy

Orbi picks it up and writes the code in an isolated working copy.

02A second session reviews it
PRReviewFixAcceptance

It checks the PR against the Issue's acceptance criteria and fixes what fails.

03Merge and release
Reviewed commitMergeTag

Only the reviewed commit merges, then Orbi tags the release.

  • Open source (AGPL-3.0), free forever
  • Self-hosted — code never leaves your machine
  • Bring your own model

One Issue: reviewed,
merged, and released

Issue #48 entered the real Orbi repository. An isolated run implemented it, a separate session reviewed the exact diff, PR #193 merged, and the work shipped in Release v0.2.0 with every Issue in scope verified item by item. The record is still inspectable because the record is GitHub.

  • 01Independent review that can fix

    The reviewer changes code, reruns the suite, and judges the repaired head.

  • 02Merge evidence tied to one SHA

    A fresh base, a mergeable PR, and the exact commit that received the verdict.

  • 03Recovery that never starts over

    Every step is written back to the Issue and the pull request, so a restart continues where the last one stopped.

Issue #1018 · PR #1023 · Release v0.5.17

Run this on your own repository

curl -fsSL https://aiready.sh | sh

Needs a machine that stays on — Linux with systemd, or macOS with launchd — plus git, gh, curl and a model endpoint. The installer adds the Python runtime and the coding agent; orbi doctor names whatever is missing.

Prefer to run the four steps by hand
git clone https://github.com/orbi-build/orbi.git && cd orbi
uv tool install --force --reinstall --editable --python /usr/bin/python3 .
cp .orbi.example.toml orbi.toml
orbi setup --config orbi.toml
Try Orbi Cloud free → Read the setup guide

Prefer to self-host? Run Orbi on your own machine — the open source (AGPL-3.0) core is free forever. Setup guide ↗

ORBI BUILDING ORBI · LIVE Inspect the public records →

orbi-build/orbiOrbi itself

79days running
1230issues closed
857PRs merged
143releases shipped

orbi-build/orbi-websiteThis website

72days running
494issues closed
795PRs merged
800deploys shipped

orbi-build/orbi-cloudPrivateCloud service

72days running
1854issues closed
1347PRs merged
203releases shipped

Teams already running Orbi

GitHub contributor iM5LBGitHub contributor Amroashraf3500GitHub contributor ansalhkayat-collabGitHub contributor masacretoolsGitHub contributor hitesh-neevedGitHub contributor geodashtops2-lgtmGitHub contributor mbuyablessing2026-langGitHub contributor dilgithxx91-sudoGitHub contributor abbasali198282GitHub contributor lutfianjay1122-createGitHub contributor benzfidrian-ship-itGitHub contributor claytonalvesrx-droidGitHub contributor mortezahatami2026-stackGitHub contributor usdeshaGitHub contributor samanarajesh-devGitHub contributor novpriyan22-svgGitHub contributor fake200ss0GitHub contributor bakegoodkofi-art+40

Try a few merged pull requests free, then pick the plan that fits.

First 3 merged pull requests free · No credit card required

SOLO

Solo

US$29 / month

  • 400M tokens included
  • 1 repository

Founding partners: 50% off for life. 5 of 6 places left.

See full pricing →

Questions people ask
before running an agent on their repo

01What does Orbi actually need to run?

For self-hosted Orbi you need a machine that stays on — Linux with systemd, or macOS with launchd — plus git, gh, curl, and API access to a model you already pay for. The installer adds uv, the Python runtime and the coding agent; if the coding agent is not installed yet, bring Node 22.19 or newer with npm. No GPU is required — the model does the thinking, and it can be a hosted API. The setup guide walks through the rest.

Self-hosted: your repository, keys, and test suite stay on your machine; nothing is uploaded to us. Cloud: the hosted worker clones the repository you authorize onto machines we operate, for as long as a delivery runs — see “Can you see my code?” on the Cloud page.

02Which AI models can I use?

For self-hosted Orbi, bring your own model. Orbi drives the agent through your credentials, using any OpenAI-compatible API or a locally hosted model, and you can change your mind later without changing your workflow.

Model spend goes to your provider at your rate. Orbi charges nothing on top: the self-hosted core has no platform fee. Cloud includes model usage.

03How is this different from Copilot or Cursor?

Those are editors — they make you faster while you are typing. Orbi is what happens when nobody is typing: it takes an Issue from your backlog, works in an isolated working copy, and takes it through review and merge to a tagged release.

The difference that matters is what happens after the code is written. A second session reviews the exact diff, repairs it, and reruns the suite, and only the reviewed commit is allowed to merge. That review step is the product.

04Who reviews the AI's code?

Another Orbi session, running independently of the one that wrote the code, and it can change the branch rather than only comment on it. It reruns the tests against its own repairs and judges the repaired head.

You stay the last gate. The work lands as a normal GitHub pull request, so your branch protection, required checks, and human approvals apply exactly as they do today.

05What happens when a run crashes halfway?

Nothing has to be explained twice. Because the Issue, the pull request and the labels live in GitHub rather than in a process, a killed process or a failed test cannot leave two versions of what happened.

A restart rebuilds its state from that record: the same branch, the same pull request, the same work in progress.

06Is it really free, and what is Managed Cloud?

Open source under AGPL-3.0, or the Sustainable Use License if your policy rules out AGPL. Self-hosting is free under either.

Managed Cloud starts with GitHub: sign in, install the App. No credit card or subscription required. Paid annually, Solo is US$24/month (US$290/year, save 17%; monthly billing US$29) with 400M tokens and Pro is US$66/month (US$790/year, save 17%; monthly billing US$79) with 1.2B tokens. Founding partners get 50% off forever, limited to 6 places.

Try it on your own repository

Try Orbi Cloud on your own repository, then let Orbi carry the work to a tagged release.

One real delivery, every week

One email a week about one real delivery. Unsubscribe anytime.