PipeCD
What it is
PipeCD is an open-source, GitOps-style continuous delivery platform for deploying and operating applications consistently. It runs deployment operations through Git pull requests and provides a shared interface for Kubernetes, Terraform, GCP Cloud Run, AWS Lambda, AWS ECS, and other targets. A plugin architecture lets users add deployment support for custom applications or platforms. PipeCD says deployments need a pipeline definition and application manifests, without requiring changes to CRDs or application manifests. Built-in deployment analysis evaluates impact using metrics, logs, and emitted requests. The platform can automatically roll back after an analysis or pipeline-stage failure, detect configuration drift, and trigger deployments from defined events. Its interface shows deployment pipelines, per-deployment logs, real-time application state, and delivery-performance insights. PipeCD is self-hosted: a Control Plane manages deployments centrally, while piped agents execute them using plugin binaries. Security features include single sign-on and role-based access control; credentials stay out of the control plane, and piped makes only outbound requests. For private Git repositories, piped requires a private SSH key and a corresponding public key added to the repository. PipeCD is available under the Apache 2.0 license.
For and against
For
- Git pull requests drive deployment operations.
- One interface supports Kubernetes, Terraform, and listed cloud targets.
- Automatic rollback can follow analysis or pipeline-stage failure.
- Detects configuration drift and can trigger event-based deployments.
- Open source under the Apache 2.0 license.
Against
- PipeCD is self-hosted, requiring users to run its components.
- Private Git repositories require matching SSH keys.
- Deployment support for custom targets uses plugins.
The verdict
PipeCD combines Git-based deployment workflows with analysis, rollback, drift detection, and deployment visibility. Its self-hosted model and private-repository key requirement are important operational considerations.
Who it suits
PipeCD suits teams that want to manage continuous delivery through Git workflows and operate deployments on their own infrastructure. It supports teams deploying across the listed cloud, infrastructure, and container targets.
Plans and prices
Open source
Apache 2.0 license
Read 1 Oct 2026 · pipecd.dev
On the record
- Analysis integrations
- Named analysis integrations include Prometheus, Datadog, Stackdriver and CloudWatch.pipecd.dev · read 15 Oct 2026
- API
- The control plane provides gRPC connectivity for piped agents and web/API endpoints for clients.pipecd.dev · read 15 Oct 2026
- Architecture
- The control plane is a web service exposing UI, API endpoints and metadata storage, while piped is a lightweight agent that runs plugins.pipecd.dev · read 15 Oct 2026
- Automatic rollback
- The product can stop a failed deployment and roll back to the preceding state.pipecd.dev · read 15 Oct 2026
- Automation
- PipeCD can automatically roll back after an analysis or pipeline-stage failure, detect configuration drift and trigger deployments from defined events.pipecd.dev · read 1 Oct 2026
- CNCF code metrics
- CNCF reports 176 GitHub stars, 239 forks and estimated software value of $17.7 million.cncf.io · read 15 Oct 2026
- CNCF contributors
- CNCF reports 494 contributors and 118 contributing organizations for PipeCD.cncf.io · read 15 Oct 2026
- CNCF health score
- The CNCF project page shows a health score marked Concerning (33).cncf.io · read 15 Oct 2026
- CNCF project age
- CNCF lists PipeCD's first commit as March 9, 2020.cncf.io · read 15 Oct 2026
- CNCF status
- CNCF accepted PipeCD as a Sandbox project on May 17, 2023.cncf.io · read 15 Oct 2026
- Control-plane storage
- The architecture supports managed or self-managed MySQL-compatible data storage and managed or self-managed object/file storage such as S3 or Minio.pipecd.dev · read 15 Oct 2026
- Credential handling
- PipeCD states that deployment credentials remain in the cluster and are not stored in the control plane.pipecd.dev · read 15 Oct 2026
- Deployment analysis
- PipeCD can evaluate deployments with metrics, logs and emitted requests through analysis providers.pipecd.dev · read 15 Oct 2026
- Deployment targets
- The same deployment interface supports Kubernetes, Terraform, GCP Cloud Run, AWS Lambda, AWS ECS and more.pipecd.dev · read 1 Oct 2026
- Documentation
- Documentation includes overview, concepts, quickstart, installation, user and contributor guides, examples, feature status, FAQ and release links.pipecd.dev · read 15 Oct 2026
- Drift detection
- PipeCD detects configuration drift and can notify users about changes.pipecd.dev · read 15 Oct 2026
- Event triggers
- Deployments can be triggered by events such as a container image push or Helm chart publication.pipecd.dev · read 15 Oct 2026
- GitOps workflow
- PipeCD enables deployment operations through pull requests on Git.pipecd.dev · read 1 Oct 2026
