monkCapsules.
Every branch gets
a place to run.
Give each branch a running application in your cloud. Frontend, API, workers and the services they depend on—ready for a person or an agent to test.
Open the branch’s web app, exercise its API and inspect the services behind it.
Example topology. Choose what each capsule creates and what it connects to when you review the plan.
The whole application.
The boundaries you choose.
A preview needs more than a frontend URL. Capsules use your application’s deployment definition to bring up its services and connections. You decide how much infrastructure each branch needs.
Dedicated cloud capacity
Provision capacity for each branch in the selected provider and region. Use separate resources where the change needs its own environment.
For infrastructure changes and independent environments.A shared cluster
Run branch workloads on an existing cluster and its selected node pool. Each capsule has its own deployment lifecycle while sharing the underlying capacity.
For parallel development on infrastructure you already have.Make the data boundary explicit.
A separate environment does not automatically copy production data. Choose fresh databases, seed data or approved shared services in the plan. A shared database also means shared writes; keep destructive tests on isolated data.
A lifecycle that
follows the work.
Set up the policy once. Keep the deployment definition and workflow in your repository, where the team can review them.
Read the Capsules documentation- 1
Set up once
Ask Monk to set up Capsules. Review the cloud or cluster, branch filters and required secrets, then commit the generated GitHub Actions workflow.
- 2
Push and review
An eligible branch provisions its environment and deploys your app. Further pushes update it. Use its preview URL, endpoints and logs to test the change.
- 3
Schedule its hours
Set a project schedule and override it for an individual capsule. Keep a demo available, or take development environments down outside working hours.
- 4
Close the branch
The generated workflow handles cleanup when the pull request closes. Check capsule status to confirm teardown and investigate any failed resources.
Before the first branch.
What do I need to get started?
A GitHub repository, a Monk deployment definition for the app and access to the target cloud or cluster. Monk prepares the Capsules workflow and requests the credentials and setup approval it needs. You review and commit the generated workflow.
Can I keep main and release branches out?
Yes. Configure branch inclusion and exclusion rules for the project. Capsules are intended for the branches you choose; the workflow does not need to replace your production deployment process.
What happens to data when a capsule goes down?
That depends on the resource and its lifecycle configuration. Plan which data is disposable, which must persist and how it will be restored before scheduling shutdowns or cleanup. Hibernation is not a backup.
Who pays for the infrastructure?
Resources run in your account and your provider bills you for them. Separate branch environments add capacity and managed-service costs. Schedules help control when environments run; stored data and other retained resources may still incur charges.
Give the next change
somewhere to prove itself.
Start with one branch. Review its environment, deploy the app and test the whole change.
Set up Monk Explore parallel agent workflows