Frequently Asked Questions

KubeKanvas is an end-to-end, AI-powered Kubernetes management platform. It is a full Kubernetes IDE that lets you design infrastructure visually, collaborate with your team in real time, deploy directly to any Kubernetes cluster (local, on-premises, or cloud), monitor live release status, and debug workloads — all from a single browser-based interface.

With a built-in AI co-pilot called Kaygent, browser-side secret encryption, multi-cluster support, and production-ready templates, KubeKanvas covers the entire lifecycle of Kubernetes workload management. See every capability in detail, or how a project flows end to end.

KubeKanvas is designed for DevOps engineers, platform engineers, and developers who work with Kubernetes. It is particularly helpful for teams that want to visualize their Kubernetes resources and relationships, streamline the creation of complex configurations, and reduce the learning curve for new team members.

It also suits developers who ship to Kubernetes without wanting to specialize in it, security-conscious users who prefer to design offline and export YAML, and IT managers who need visibility and standardized deployment practices across several teams.

KubeKanvas gives you a visual canvas to design Kubernetes resources using drag-and-drop, with an integrated AI co-pilot that can suggest, create, and configure resources alongside you. You can start from scratch, from a production-ready template, or by importing YAML you already have.

Once your design is ready, connect a cluster with the KubeKanvas CLI — local (k3d, minikube, Docker Desktop), on-premises, or cloud-hosted — and deploy your project directly with a few clicks. From there, the live release monitor tracks your deployment in real time, showing workload health and resource status as it rolls out.

You can install new releases, upgrade existing ones, and troubleshoot issues without leaving the platform using built-in Describe, Logs, and YAML views. You can also export standard YAML or a Helm chart for your CI/CD pipelines, and manage multiple clusters from one unified interface.

KubeKanvas addresses the full spectrum of challenges teams face when working with Kubernetes:

• Eliminates manual YAML authoring with a visual drag-and-drop designer and AI assistance
• Reduces context switching by combining design, deployment, monitoring, and debugging in one platform
• Enables direct deployment to local, on-premises, and cloud Kubernetes clusters without extra tooling
• Provides real-time release monitoring so teams can catch and resolve rollout issues immediately
• Protects sensitive values with browser-side secret encryption — secrets never leave the browser in plaintext
• Lowers the Kubernetes learning curve with production-ready templates and AI guidance
• Improves team alignment by making infrastructure designs a shared, visual source of truth for architects, developers, and DevOps
• Simplifies multi-cluster management across different cloud providers from a single unified interface

There are three common approaches: generate a static diagram from your manifests, use an in-cluster dashboard to inspect what is already running, or import the manifests into a visual editor that maps resources and the references between them. The first two show you a snapshot; only the third lets you change anything.

KubeKanvas takes the third approach. Paste or upload existing YAML and it is laid out on a canvas, with Services, Ingresses, ConfigMaps, Secrets, and volumes drawn against the workloads that reference them, so you can see what depends on what. Because the canvas is editable, the diagram is the configuration rather than a picture of it — adjust a resource visually and the generated YAML changes with it. See how it works.

For a single resource, kubectl create ... --dry-run=client -o yaml is the quickest way to get a valid skeleton to edit. The difficulty starts when you need a realistic set — a Deployment plus its Service, Ingress, ConfigMap, Secret, and PersistentVolumeClaim — because that is where label selectors stop matching, indentation drifts, and small mistakes only surface at apply time.

KubeKanvas generates the whole set from one visual design, validating against Kubernetes schemas and best practices while you edit, and outputs both standard YAML and a Helm chart you can apply or commit unchanged. The built-in Kaygent AI co-pilot can scaffold resources from a plain-English description of what you want to run, and asks for approval before it changes a design. Browse the features.

Yes. KubeKanvas generates a Helm chart from every design, so the canvas doubles as a chart editor: add resources and fill in values through forms, and the chart templates and values files are produced for you.

You can deploy the chart straight to a connected cluster from the browser, or pull it locally with kubekanvas fetch-helm-chart and run helm install or helm upgrade yourself. For pipelines, the helm-kubekanvas post-renderer decrypts encrypted values at deploy time so no plaintext secrets pass through CI. See Helm deployment automation.

It depends which half of the job you mean. Cluster-inspection GUIs are good at showing you what is currently running and letting you poke at it, but they are generally not built for authoring the configuration in the first place — so you end up back in an editor writing manifests by hand.

KubeKanvas covers both halves: design workloads on a canvas, deploy them to a connected cluster, then watch the release in real time and run Describe, fetch workload logs, and inspect live YAML from the same interface — including across multiple clusters on different providers. It runs in the browser, with a small CLI agent handling the cluster connection.

Mechanically it is the same as any other cluster: point your kubeconfig context at the local cluster and apply your manifests with kubectl apply or helm install. The only real difference is that the API server is running on your own machine, so image pulls and ingress usually need a little extra setup.

With KubeKanvas, install the CLI and run kubekanvas login, kubekanvas configure, and kubekanvas connect to register the cluster — k3d, minikube, and Docker Desktop all work — then deploy from the browser in a few clicks. If you do not have a local cluster yet, kubekanvas setup-k3d provisions a k3d one for you. Direct deployment is available on every plan, including the free tier. Command reference.

The thing to know first is that Kubernetes Secrets are only base64-encoded, not encrypted, so a manifest containing one should never be committed to Git as it stands. The two usual answers are to keep the values in an external secret manager and reference them at deploy time, or to encrypt the values themselves so the encrypted manifest is safe to commit.

KubeKanvas takes the second approach: Secret values are encrypted in your browser with your workspace public key, using AGE, before they leave the device. Encrypted values carry an AGE-ENC: prefix and are safe to commit; the matching private key stays on the machine running the CLI and is used to decrypt at deploy time. KubeKanvas servers, its database, and Kaygent only ever see ciphertext. How Secret encryption works.

The usual progression is templating first — Helm or Kustomize, to stop duplicating the same Deployment across environments — and then a shared library of known-good starting points, so that nobody on the team begins from a blank file.

KubeKanvas does both visually. Start from a production-ready template or import YAML you already have, adjust resources through forms with validation as you type, and save your own templates so internal patterns stay consistent across projects. Kaygent can draft a first version from a description of the workload. You still get standard YAML and a Helm chart as output, so nothing about what you produce is locked to the platform.

Yes, KubeKanvas allows you to import existing YAML files. You can upload your YAML configurations, and KubeKanvas will visualize them on the canvas. This makes it easy to understand, modify, and extend your existing Kubernetes configurations, so the platform works with clusters that were not built in it.

Yes, KubeKanvas supports Kubernetes Custom Resource Definitions (CRDs). You can create, visualize, and manage custom resources alongside standard Kubernetes resources, including Gateway API types and cert-manager Certificates and Issuers. This is particularly useful for teams using operators or extending Kubernetes with custom functionality.

Yes. Projects are shared, updates appear live for everyone in the project, and version history plus in-canvas comments mean review happens where the design is rather than in a YAML diff. Every action, human or AI, is undoable and redoable.

The practical effect is that a design becomes a single visual source of truth that architects, developers, and DevOps engineers can all read, instead of a set of manifests passed around as files. More on collaboration.

Security is the reason several parts of KubeKanvas are built the way they are. KubeKanvas never needs inbound access to your cluster: the KubeKanvas CLI runs inside your own network, opens an outbound connection to the platform, and receives deployment instructions rather than credentials. Deployments execute under the kubeconfig and RBAC permissions you already hold, so clusters behind a firewall or NAT work without opening anything up, and KubeKanvas cannot exceed the access you have yourself.

Secret values are encrypted in your browser with your workspace public key before they are stored anywhere. KubeKanvas servers, its database, and the Kaygent AI co-pilot only ever see AGE ciphertext or [ENCRYPTED] placeholders, and the matching private key stays on the machine running the CLI. See Secret encryption for the full key-pair model.

Access is managed through Keycloak, so multi-factor authentication, role-based access control, SSO, and session management extend from your existing identity provider into KubeKanvas — including CLI logins, which use device flow rather than a stored password. Project data is encrypted in transit and at rest. Full detail is on the security page.

Yes, there is a free tier that covers individuals and small teams, and it includes direct cluster deployment rather than export-only. Paid plans add higher usage limits, team features, and enhanced support. Current plans and prices are on the pricing page — no credit card is needed to start.

Yes, we offer enterprise plans for organizations with specific requirements. Enterprise plans include SSO integration, dedicated support, custom integrations, advanced security features, and an on-premises deployment option. Contact us for enterprise pricing and customization options.

Yes, you can upgrade or downgrade your plan at any time. When upgrading, you will be charged the prorated amount for the remainder of your billing cycle. When downgrading, the new rate will apply at the start of your next billing cycle. You can manage your subscription from your account settings.

Yes, we offer special pricing for non-profit organizations, educational institutions, and open-source projects. Get in touch with details about your organization to learn more about our discount programs.