Spacelift is a control plane that executes runs on workers. You can use shared workers or run private worker pools in your cloud for full data path control, with end-to-end encryption of run payloads. If your repos are private to the internet, a lightweight VCS agent lets Spacelift reach them securely.
A stack is an isolated unit that combines your IaC repository, current state, and configuration, such as environment variables and mounted files. It is the primary entity you operate on in Spacelift.
Spacelift supports Terraform, OpenTofu, Terragrunt, Pulumi, AWS CloudFormation, Kubernetes, and Ansible, so you can standardize workflows across mixed tooling.
A run is a job that can preview or change infrastructure. Common types are proposed runs for safe previews and tracked runs for actual deployments, with optional approvals in between.
Spacelift detects drift by executing scheduled proposed runs against finished state and flagging any differences. You can review findings in the UI and optionally reconcile, and it works on private workers.
Connect your Git provider and select the repo. Spacelift supports GitHub, GitLab, Bitbucket, and Azure DevOps, including self-hosted and SSH.
Use Contexts to share environment variables, mounted files, and hooks across stacks and modules. Attach a context wherever you need it to avoid copy-pasting configuration.
Spacelift uses Open Policy Agent with Rego to enforce rules at every stage. You write readable policies that check plans, control who can approve or apply, govern which stacks can touch which accounts, and block risky changes before they land. Policies evaluate in the worker context with full input data, return clear messages to engineers, and become part of your compliance record.
Yes. There is a self-hosted edition that you install in your own infrastructure with provided images and templates for clouds like AWS, GCP, and Azure, as well as on-premise Kubernetes environments.