Atlantisvs
HCP Terraform


HCP Terraform: Managed runs, state and policy with a vendor contract, when self-hosting the webhook server is unacceptable

Runs Terraform plan and apply from pull request comments, self-hosted and free
As of 31 August 2026, Atlantis is free to use. Atlantis watches pull requests, runs the plan and posts the output as a comment so review happens against the actual diff rather than a description of it. Softwr lists it under Developer Tools. Atlantis is made by Atlantis (community project under LF Projects, LLC), available on Linux, Docker, Kubernetes, Self-hosted.
Overview
Atlantis is a self-hosted Go application that receives webhooks from GitHub, GitLab, Bitbucket or Azure DevOps, runs Terraform or OpenTofu operations on a server you control, and writes the results back to the pull request. Developers then type atlantis apply as a comment to execute. Because plan output lands in the review thread, the change a reviewer approves is the change that runs, which removes the most common failure in Terraform review: approving a description of a plan somebody ran locally. It supports workspace locking, autoplanning, automerging and Conftest policy checks. The governance is unusual and worth stating plainly. There is no company. The project operates as a series of LF Projects under the Linux Foundation umbrella and is powered by volunteer contributions. It is not a CNCF project, despite using a channel in the CNCF Slack, which is easy to misread. The maintainer file lists six people employed at six different and unrelated organisations, which is a genuinely better continuity position than a single-vendor open source tool where the vendor can relicense or pivot. Against that, there is no GOVERNANCE file in the repository, so succession and decision-making are informal, and the repository shows around 9,300 stars, over 400 contributors and version 0.47.1 released in August 2026. Buyers are infrastructure teams that want pull request driven Terraform without a per-user subscription and are willing to run and secure the server. The trade-off is that this component holds privileged cloud credentials on an internet-reachable webhook endpoint, and authorisation is driven by pull request comments, which is coarse compared with a policy engine. It is also stateful, holding a working directory and lock database on local disk, which makes high availability awkward. Compared with commercial alternatives it has no drift detection, no cost estimation and no managed state interface, and Terragrunt or monorepo layouts require hand-written workflow configuration that grows with the repository count.
The honest half
Concrete and checkable, so you can decide whether any of them matter to you. This is the half of a review a vendor will not write about Atlantis.
Cross-shopped
Each pairing was judged by two reviewers asking whether a buyer would genuinely weigh the two against each other. The ones that failed were deleted rather than published.


HCP Terraform: Managed runs, state and policy with a vendor contract, when self-hosting the webhook server is unacceptable


Harness: Broader delivery platform if infrastructure pipelines sit alongside application deployment


Pulumi: Choose this instead if you would rather define infrastructure in a general purpose language
Pricing
Taken from the vendor's own pricing page. Prices move, so check before you buy.
Atlantis
Free
Capabilities
Pull request plans
Runs plan automatically when infrastructure code changes and posts output as a comment
Comment driven apply
Executes apply when an authorised reviewer comments on the pull request
Workspace locking
Prevents two pull requests planning against the same state at once
Policy checking
Conftest integration so policy failures block an apply
Multi-platform webhooks
Works with GitHub, GitLab, Bitbucket and Azure DevOps
Custom workflows
Per-repository configuration for non-standard directory layouts and tooling
Answered, with sources
Each answer names the page it came from, so you can check it rather than take our word for it.
No. It runs as a series of LF Projects under the Linux Foundation and is maintained by volunteers. There is no vendor to buy support from and no service level agreement.
No. It uses a channel in the CNCF Slack, which is often misread as membership, but the project sits under LF Projects rather than the CNCF.
Yes, there is a dedicated integration guide alongside Terraform, and it also has configuration for working with hosted Terraform backends.
Drift detection, cost estimation, a managed state and run interface, and a real policy engine. Atlantis does pull request plan and apply well and stops there.
Keep looking
Infrastructure-as-code orchestration for Terraform, OpenTofu, Pulumi and Kubernetes
Infrastructure as code provisioning and management, formerly Terraform Cloud
Software delivery platform across CI, CD, security testing and cost management
Open source event pipeline that streams behavioural data to your own warehouse
Local Kubernetes development loop that rebuilds and live-updates containers on save
Self-hosted, Pusher-compatible WebSockets server with a stalled maintenance record
Fast, scalable build system with intelligent defaults for Python and more
Query cloud APIs, SaaS tools and code with SQL, with no extract or load step
Extensible, customizable free text editor and computing environment
Runs production-like development environments on your own Kubernetes cluster
Internal developer portal that scores service ownership and production readiness
Platform for building and deploying realtime multiplayer servers on the edge
AI-powered memory layer that captures and organizes developer work
Cross-platform native app development with Delphi and C++Builder
Softwr does not host reviews and shows no star rating for Atlantis, because a rating we did not collect is not ours to publish. What is here is the pricing and platform detail from the vendor’s own pages, limitations we could state concretely, and alternatives a reviewer confirmed people weigh against it. Tell us if any of it is wrong.
What people switch to, and what they give up
Every tier, and where the cost actually lands
Put it head to head with anything we hold
Its rating, and an embed for your own site
Internal developer portal that scores service ownership and maturity rather than just cataloguing services
quoteFeature flag service priced per flag and environment with unlimited seats and unlimited monthly active users
Per month by feature flag and environment countProgrammable CI/CD engine that runs your pipeline as containers on any runner
Per month per team for Dagger CloudEvent gateway that receives, queues, retries and replays inbound webhooks
Per month by delivered event countAnalytics governance: define a tracking plan, generate typed code from it, and catch drift before release
Per editor seat per monthKubernetes development and testing automation with shared caching, now owned by Incredibuild
Open source, no licence feeThe Python web framework behind ERPNext, sold as managed hosting through Frappe Cloud
Per month by site or serverOpen source internal developer portal framework created and open sourced by Spotify
Open source, no licence fee