Stackorder vs Terramate: Terraform and OpenTofu stacks in dependency order
In short
- Both
- Run Terraform and OpenTofu on CI runners you control, order stacks within a repository, plan every stack that uses a changed local module, and hold no cloud credentials, in Stackorder's case by default.
- Terramate
- An MPL-2.0 CLI with no server, ordering from
afterandbefore, code generation and blueprints for several CI systems; Terramate Cloud, a SaaS with a free plan, adds pull request previews, drift and dashboards. - Stackorder
- A dependency graph that plans every downstream stack on each pull request and applies in waves on your GitHub Actions runners. Apache-2.0, with a small server you host; GitHub and S3 only.
Terramate and Stackorder both plan and apply Terraform and OpenTofu stacks in dependency order, on CI runners you control, and neither runs Terraform on a server of its own. Terramate is an MPL-2.0 CLI: a stack is a directory with a stack {} block, its after and before attributes set the order, and --changed selects the stacks a Git change touches. Terramate Cloud collects the plans, deployments and drift checks your CI syncs to it.
Stackorder keeps the dependency graph on a small server you host. It plans every stack a change reaches, applies the affected stacks in waves from a pull request comment or on merge, and keeps stack locks and run history in Postgres.
Stackorder and Terramate side by side
Numbers link to the sources. A dash means we haven't verified it, not that it's missing.
| Feature | Stackorder | Terramate |
|---|---|---|
| License | Apache-2.0, open source41 | MPL-2.0 for the CLI; Terramate Cloud is a commercial SaaS2, 4 |
| Deployment | Self-hosted; setup mode creates the GitHub App from a manifest32, 37 | The CLI runs in your CI or shell; Terramate Cloud is SaaS, or a Private Cloud instance on Enterprise4, 5 |
| Pricing | Free and open source; you run the server41 | CLI free. Cloud Community: free, 2 users, 1,000 resources. Teams: $449 a month, unlimited users, 5,000 resources. Enterprise: custom4 |
| Maturity | v0.6.0, released 2026-11-15; first released as v0.1.0 on 2026-09-3039, 40 | Actively released: v0.17.3 on 2026-09-08, four releases since May 20263 |
| Where Terraform runs | Your GitHub Actions runners, GitHub-hosted or self-hosted; it manages no runners or agents28 | Wherever the CLI runs: it runs any command in each selected stack, in your CI or locally; Terramate Cloud runs nothing5, 8 |
| State backend | Bring your own S3; never takes or releases the state lock28 | Your existing backend; Terramate Cloud needs no access to it5 |
| Modules | No registry; lists each module's consumers, and for git modules the version each pins and how far behind it is28 | No registry; change detection follows local modules referenced by relative path, through any depth of nesting12 |
| Self-hosted footprint | One container of about 30 MB and Postgres; actions that use no Docker28 | No server to self-host: the CLI in your CI, with Terramate Cloud as SaaS5 |
| Cross-stack dependencies | A graph of stacks and modules from depends_, module sources, and terraform_ and stackorder_ reads, including cross-repository edges; applies in waves29, 30, 36 | Parents run before children, and after and before order stacks; these order a run but add no stacks to it. --include- adds stacks that read outputs or Terragrunt dependency blocks7, 9, 10, 11 |
| Sharing outputs between stacks | Shared outputs, off by default: the producer lists its readers in outputs., and each reader uses the stackorder_ data source or ephemeral resource. The server returns only the outputs, sensitive ones only if the producer shares them, so the reader's role needs no access to the producer's state; a reader's plan fails while the server is unreachable31, 36 | Outputs sharing, experimental: an input block reads another stack's output at run time through a command, usually terraform output -, which reads that stack's state; mock values cover plans before it applies14, 26 |
| Cloud credentials | Not held by the server: each job assumes your IAM role with its own GitHub OIDC token. Shared outputs, off by default, give the server a read-only role on the state objects you list31 | Not held: the CI job assumes your role, with OIDC in the blueprints; Terramate Cloud needs no access to cloud accounts5, 18 |
| Human sign-in | GitHub OAuth through the App, read:org scope only; the AWS Terraform module can also require OIDC or Cognito sign-in at the load balancer, in front of the UI32, 38 | Terramate Cloud: SSO on every pricing tier; SAML 2.0, RBAC and audit logs on Enterprise4 |
| Git hosts | GitHub only, by design28 | CI blueprints for GitHub Actions, GitLab CI and Bitbucket Pipelines; pull request previews sync from GitHub Actions only16, 20 |
| OpenTofu | Yes, with tool: tofu; tested end to end with OpenTofu 1.1235, 39 | Yes; it runs tofu like any other command, and Terragrunt too24, 25 |
| Drift detection | Scheduled per stack with drift.; with open_, one GitHub issue per drifted stack, closed when the drift is gone; never applies to fix drift33 | Drift workflows you schedule in your CI, synced to Terramate Cloud, with Slack notifications21 |
| Policy checks | Not a policy engine; run OPA, conftest, Checkov or Infracost in hooks, and stackorder check records a named check the apply gate honors34 | Terramate Cloud: more than 500 built-in CIS Benchmark policies, run in drift checks, and policy as code with OPA4, 22 |
| Pull request workflow | A check per stack, one sticky comment, and stackorder plan, apply and unlock comments; applies before merge by default, or on merge29 | The blueprints plan changed stacks on each pull request and apply them on push to main; Terramate Cloud posts a preview summary in the pull request, with the full plans in Terramate Cloud17, 18, 19 |
Key differences
Where the order lives
Terramate works out the order in the CLI on every run, from the directory hierarchy and each stack's after and before, and needs no server. Stackorder's server stores the graph, takes stack locks and dispatches applies one wave at a time. When it is down, applies are refused while plans still run, except those of stacks that read shared outputs.
What a change plans
A change in Stackorder plans the changed stacks, every stack that uses a changed local module, and, by default, their dependents over depends_on and state reads, so reviewers see the downstream effect before anything applies. Dependents whose plan is a no-op are skipped at apply time.
Sharing outputs
Terramate's outputs sharing is experimental. A consumer's input runs a command such as terraform output -json in the producing stack, which reads the producer's state, so the job needs access to it; mock values stand in when both change in one pull request. With Stackorder's shared outputs, the producer lists its readers and the server returns only the outputs, so a reader's role needs no access to the producer's state.
Pull requests and applies
Terramate's GitHub Actions blueprints plan changed stacks on every pull request and apply them on push to main, and Terramate Cloud posts a preview summary in the pull request, with the full plans in Terramate Cloud. Stackorder posts a check per stack and one sticky comment, applies from a stackorder apply comment before merge by default or on merge, and refuses an apply while another pull request holds a lock on an affected stack.
Across repositories
Terramate Cloud can connect several repositories to one organization, and the CLI orders stacks within one project. Stackorder records edges between stacks in different repositories; when an upstream stack applies, it can start plan-only runs in the repositories downstream, though each run applies one repository.
Scope and setup
Terramate also generates HCL and other files across stacks, runs any command, works with Terragrunt and any state backend, and has blueprints for GitLab CI and Bitbucket Pipelines. Stackorder is GitHub only, and needs a server and Postgres that you host.
Where Terramate is strong
- No server to run: the CLI orders and runs commands across stacks wherever you call it.5, 8
- Change detection that follows local modules through any depth of nesting.12
- Change detection for Terragrunt
include,dependencyanddependenciesblocks, andterramate create -to turn an existing Terragrunt project into ordered stacks.13, 25-all- terragrunt - Code generation of HCL and other files across stacks, with globals.15
- Runs any command per stack, so Terraform, OpenTofu and Terragrunt all work.8, 24
- CI blueprints for GitHub Actions, GitLab CI and Bitbucket Pipelines.16
- Terramate Cloud needs no access to your repositories, state backend or cloud accounts, and the CLI sanitizes plans before it syncs them.5, 6
- A free Terramate Cloud plan with pull request previews and drift management, for 2 users and 1,000 resources.4
When to choose which
Choose Stackorder when
- Your code is on GitHub and you want Terraform or OpenTofu to run on your own GitHub Actions runners, under AWS roles the runner assumes with its own GitHub OIDC token.
- You have many stacks that depend on each other or on shared modules, and you want a change planned everywhere it lands and applied in dependency waves.
- You want a small open-source server you host yourself, which stores no state. By default it has no cloud access, and an outage pauses applies but not pull request plans.
Choose Terramate when
- You want ordering and change detection with no server to run, from a CLI that works in any CI or on a laptop.
- Your code is on GitLab or Bitbucket, or your state is in a backend other than S3.
- You use Terragrunt, or want code generation across stacks.
- You want hosted dashboards, pull request previews and drift reports rather than a server you host.
- You want a tool with more production use: Stackorder's first release, v0.1.0, came out on , and the current version is v0.6.0.
Try Stackorder on your own repositories
Free and open source under the Apache License 2.0. The getting started guide takes one repository from nothing to a first stackorder apply; the local demo runs on one machine with no GitHub App and no AWS account.
Frequently asked questions
Is Stackorder an alternative to Terramate?
How does Terramate order stacks?
after and before attributes, which take stack paths or tag filters, set the rest of the order. terramate run --parallel runs independent stacks at the same time and keeps that order. Stackorder orders stacks by depends_on and by terraform_remote_state and stackorder_outputs reads, and applies them in waves.Does Terramate plan the stacks that depend on a changed stack?
--changed selects the stacks whose files or local modules changed, plus the stacks they force into the run with wants or wanted_by, which do not order them. after and before only order the stacks a run already includes. --include-all-dependents follows data dependencies only: it adds stacks that read the changed stacks' outputs through outputs sharing or Terragrunt dependency blocks. Stackorder plans the dependents of a changed stack by default.Is Terramate open source?
What does Terramate Cloud cost?
Does Terramate hold cloud credentials or state?
More comparisons
- All tools in one feature matrix
- Stackorder vs Atlantis
- Stackorder vs HCP Terraform (formerly Terraform Cloud)
- Stackorder vs Stategraph (formerly Terrateam)
- Stackorder vs Spacelift
- Stackorder vs env zero (formerly env0)
- Stackorder vs Scalr
- Stackorder vs Terrakube
- Stackorder vs OpenTaco (formerly Digger)
- Stackorder vs Terragrunt
- What Stackorder is and how it works