- Home
- AI DevOps Assistants
- K8sGPT
K8sGPT
What it is
K8sGPT scans Kubernetes clusters and explains problems in plain English. It offers a command-line interface, an in-cluster operator, and an MCP interface for AI assistants. The project lists 23 built-in analyzers and supports custom analyzers over gRPC, along with 14 cloud or local AI backends. Named backends include OpenAI, Azure OpenAI, Amazon Bedrock, Google Vertex AI, Ollama, and LocalAI; documented integrations include Prometheus, Trivy, Kyverno, KEDA, and AWS. Analysis without the --explain option does not require an AI backend, but explanations do. For local or air-gapped operation, users can choose Ollama, LocalAI, or a custom REST backend. The --anonymize option masks names and identifiers before a prompt is sent, then restores them in the answer. Operator auto-remediation is optional and subject to a configurable risk threshold. CLI, Docker, and operator installation paths require a reachable Kubernetes cluster and kubeconfig; the operator path also requires Helm 3. K8sGPT is free and open source under Apache 2.0.
For and against
For
- Offers CLI, in-cluster operator, and MCP interfaces
- Lists 23 built-in analyzers and custom analyzer support
- Can use local or air-gapped AI backends
- Free and open source under Apache 2.0
Against
- Explanations require an AI backend
- Installation paths require a reachable cluster and kubeconfig
- Operator installation requires Helm 3
The verdict
K8sGPT pairs cluster analysis with optional AI explanations and several ways to run it. Check the backend and cluster requirements for the installation path you intend to use.
Who it suits
K8sGPT suits teams troubleshooting Kubernetes clusters, including those seeking local or air-gapped AI options. Users should have a reachable cluster and kubeconfig for the listed installation paths.
Plans and prices
K8sGPT
Open source · Apache 2.0
Read 7 Oct 2026 · k8sgpt.ai
On the record
- AI backends
- The site lists 14 cloud or local AI backends, including OpenAI, Azure OpenAI, Amazon Bedrock, Google Vertex AI, Ollama, and LocalAI.k8sgpt.ai · read 7 Oct 2026
- Analyzers
- The site lists 23 built-in analyzers and says users can add custom analyzers over gRPC.k8sgpt.ai · read 7 Oct 2026
- Backend requirement
- Analysis without the --explain option works without an AI backend; explanations require an AI backend.k8sgpt.ai · read 7 Oct 2026
- Cluster access
- The installation guide requires a reachable Kubernetes cluster and kubeconfig for CLI, Docker, and operator paths.k8sgpt.ai · read 7 Oct 2026
- Integrations
- Documented integrations include Prometheus, Trivy, Kyverno, KEDA, and AWS.k8sgpt.ai · read 7 Oct 2026
- Interfaces
- K8sGPT offers a CLI, an in-cluster operator, and an MCP interface for AI assistants.k8sgpt.ai · read 7 Oct 2026
- Local operation
- K8sGPT can use Ollama, LocalAI, or a custom REST backend for local or air-gapped operation.k8sgpt.ai · read 7 Oct 2026
- Operating systems
- The installation guide lists Windows binaries, macOS and Linux Homebrew support, and Linux packages in .deb, .rpm, and .apk formats.k8sgpt.ai · read 7 Oct 2026
- Operator installation
- The guide lists Helm as the installation method for the in-cluster operator and requires Helm 3 for that path.k8sgpt.ai · read 7 Oct 2026
- Privacy
- With the --anonymize option, names and identifiers are masked before a prompt is sent and restored in the answer.k8sgpt.ai · read 7 Oct 2026
- Purpose
- K8sGPT scans Kubernetes clusters with SRE-grade analyzers and explains problems in plain English.k8sgpt.ai · read 7 Oct 2026
- Remediation
- Operator auto-remediation is opt-in and gated by a configurable risk threshold.k8sgpt.ai · read 7 Oct 2026
- Support
- The site points users to community office hours and Slack.k8sgpt.ai · read 7 Oct 2026
