Calm deploys for busy teams.
Orbit plans every release, shows you the diff, and ships it the same way every time — from a laptop or from CI. Releases you can read before they happen, and undo in seconds after.
$ orbit deploy --env production
Release #7 → production
~ service hello-api image sha256:9f2c… → 4e11…
~ env DATABASE_POOL 10 → 25
- route /v1/legacy
Apply? [y/N] y
build ok (42s)
verify ok (/healthz)
switch ok (0.3s)
Release #7 is live.
Dry-run diffs, by default
Every deploy starts as a plan a human can read. Since 0.4 the diff
isn't a flag you remember — it's how orbit deploy begins.
One config, every environment
Environments are three-line overlays on a single
orbit.yaml. Staging mirrors production because there is
nowhere for drift to hide.
Rollbacks in seconds
Every release stays addressable. Undo is a pointer flip — not a rebuild, not a redeploy, not a 40-minute pipeline at midnight.
How rollback works →