Skip to content

kubestellar/console-kb

License Part of Console Ecosystem

Community knowledge base for KubeStellar Console AI missions — share, discover, and import proven fixes to save tokens and time.

Overview

Console KB is the community hub for sharing AI mission fixes created in the KubeStellar Console. When you solve a complex Kubernetes problem using the Console's AI-powered missions, you can export and publish your solution here so others can:

  • Import proven fixes directly into their Console
  • Discover community-tested fixes to common challenges
  • Save tokens by reusing fixes instead of re-prompting AI
  • Learn from real-world multi-cluster Kubernetes fixes

How It Works

┌─────────────────────┐     Export      ┌─────────────────────┐
│  KubeStellar        │ ──────────────▶ │  Console KB         │
│  Console            │                 │  (This Repo)        │
│                     │ ◀────────────── │                     │
│  AI Missions        │     Import      │  Community Fixes │
└─────────────────────┘                 └─────────────────────┘

Fixer Format

Each fixer mission is a self-contained package that includes:

  • Mission definition — the AI prompt and parameters
  • Expected outcomes — what the mission produces
  • Prerequisites — required cluster setup, CRDs, or tools
  • Tags — categories for discovery (e.g., multi-cluster, security, networking)
  • Compatibility — Console version and tested Kubernetes versions

See CONTRIBUTING.md for the submission flow and docs/fixer-schema.yaml for the annotated format reference.

Getting Started

Browse Fixes

Explore the fixes/ directory to find community-contributed AI mission fixes organized by category.

Import a Fix

  1. Copy the fix YAML from this repository
  2. In KubeStellar Console, go to AI Missions → Import
  3. Paste or upload the fix file
  4. The mission is ready to run in your environment

Share Your Fix

  1. Create a successful AI mission in KubeStellar Console
  2. Export it via AI Missions → Export
  3. Fork this repo and add your fix to the appropriate category under fixes/
  4. Submit a PR with a description of what the mission fixes

See CONTRIBUTING.md for detailed submission guidelines.

Operational Runbooks

In addition to community-contributed fixes, this repository ships a curated runbooks/ catalog for deterministic day-2 operations. These runbooks use the same kc-mission-v1 format as fixes, but they are written for repeatable operational tasks such as upgrades, restores, certificate rotation, and RBAC remediation.

Browse runbooks/ or start with runbooks/README.md for authoring expectations and operational context.

Runbook Purpose
install-kubestellar-controller.json Install the KubeStellar core controller in standalone mode with Helm.
upgrade-kubestellar-controller.json Perform an in-place KubeStellar controller Helm upgrade with health gates and rollback readiness.
rollback-kubestellar-controller.json Roll back kubestellar-core to a known-good Helm revision after a failed or regressing upgrade.
certificate-rotation.json Rotate kubeadm control-plane certificates, refresh kubeconfig, and clear expired credential issues.
cluster-upgrade.json Upgrade a kubeadm-managed cluster with prechecks, package updates, and post-upgrade validation.
node-drain.json Cordon, drain, validate, and restore a node during planned maintenance.
rbac-audit.json Audit RBAC access and apply least-privilege remediation for RBAC_DENIED failures.
disaster-recovery.json Back up and restore kubeadm etcd state during disaster recovery for self-managed control planes.
restore-etcd-snapshot.json Restore a Kubernetes control plane from an existing etcd snapshot with rollback safeguards.
restore-velero-backup.json Restore namespaces, volumes, and workloads from a completed Velero backup.

You can import runbooks into KubeStellar Console the same way you import fixes: copy the mission file, open AI Missions → Import, and upload or paste the JSON payload.

Fix Categories

Add new fixes under the category whose scope best matches the problem you solved. If a fix clearly matches an existing category, keep related missions together there instead of creating a new top-level directory.

Category Scope Example fix type
cncf-generated/ Auto-generated fixer missions grouped by CNCF project and maintained by repository automation. A generated Alertmanager or Argo CD bug-fix mission sourced from upstream issue data.
cncf-install/ Install missions for CNCF projects. Installing cert-manager, Cilium, or another CNCF project into a cluster.
cost-optimization/ Resource efficiency, right-sizing, and spend reduction guidance. Reducing idle capacity or tuning requests and limits.
llm-d/ Missions for the LLM-D stack and related AI inference components. Installing inference scheduling, benchmark, or prefix-cache features.
multi-cluster/ Cross-cluster deployment, federation, placement, and sync patterns. Setting up multi-tenancy or propagating workloads across clusters.
networking/ Service mesh, ingress, DNS, traffic, and connectivity fixes. Repairing a NetworkPolicy or cross-namespace service path.
observability/ Monitoring, logging, tracing, and alerting workflows. Adding cluster diagnostics dashboards or alert investigation steps.
orbit/ KubeStellar Orbit operational tasks and lifecycle checks. Orbit backup verification, certificate rotation, or version-drift checks.
platform-install/ Auto-generated install missions for Kubernetes platforms, distributions, and operators. Installing Agones, CloudNativePG, or a managed-platform add-on.
security/ RBAC, secret management, policy, and vulnerability remediation. Fixing an RBAC denial or remediating a published CVE.
troubleshooting/ General diagnostic and recovery missions. Restoring expired credentials, pruning kubeconfig entries, or etcd backup checks.
workloads/ Application deployment, rollout, storage, and workload behavior fixes. Migrating workload storage or updating deployment patterns.

Category Notes

  • Choosing a category: pick the directory that matches the dominant user problem, not every tag a mission might include. For example, an RBAC error belongs in security/ even if it happens during a multi-cluster workflow.
  • General diagnostics: add new diagnostic and recovery fixes under fixes/troubleshooting/.
  • fixes/index.json: this file is auto-generated by node scripts/build-index.mjs and is refreshed by the Build Mission Index GitHub Actions workflow whenever fixes/** or runbooks/** changes land on master (excluding fixes/index.json itself). Do not edit it by hand.

Part of the KubeStellar Ecosystem

Repository Description
kubestellar/kubestellar Core multi-cluster orchestration engine (BindingPolicy, WDS, ITS, WEC)
kubestellar/console AI-powered multi-cluster Kubernetes dashboard
kubestellar/console-marketplace Community dashboards, card presets, and themes
kubestellar/console-kb (this repo) AI mission knowledge base — share and discover fixes
kubestellar/kubestellar-mcp MCP server — AI agent tooling for Claude, Cursor, Windsurf, VS Code

Contributing

We welcome contributions! Whether you're sharing a fix that saved you hours of debugging or improving an existing fix, every contribution helps the community.

Please see CONTRIBUTING.md for details on submitting fixes, local validation, and the purpose of generated repository files such as generation-report.md and search-state.json.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Community

About

Community knowledge base for KubeStellar Console AI missions — share, discover, and import proven solutions to save tokens and time

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Generated from kubestellar/repo-template