zhongwei/gh-colek42-claude-plugins-nk-devops-tools
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial commit

Browse Source
This commit is contained in:
Zhongwei Li
2025-11-29 18:15:16 +08:00
commit 8e6cf6b820
7 changed files with 603 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
.claude-plugin/plugin.json Normal file
View File

@@ -0,0 +1,14 @@
{
"name": "nk-devops-tools",
"description": "Helm, ArgoCD, Vault best practices and disk space management for Minikube/Skaffold workflows",
"version": "1.0.0",
"author": {
"name": "Cole Kennedy"
},
"skills": [
"./skills"
],
"commands": [
"./commands"
]
}

3
README.md Normal file
View File

@@ -0,0 +1,3 @@
# nk-devops-tools
Helm, ArgoCD, Vault best practices and disk space management for Minikube/Skaffold workflows

37
commands/disk-cleanup.md Normal file
View File

@@ -0,0 +1,37 @@
---
description: Clean Docker build cache in Minikube to free disk space
---
# Disk Space Cleanup for Minikube
Minikube docker build cache can grow to 200GB+ from repeated Skaffold builds. This command checks disk usage and cleans the cache without destroying cluster state.
## Check Disk Usage
```bash
docker exec minikube df -h /var/lib/docker
```
## Clean Build Cache
This frees ~240GB+ without destroying cluster state (PVCs, volumes remain intact):
```bash
docker exec minikube docker builder prune -af && docker exec minikube docker image prune -af
```
## What Gets Cleaned
- Docker build cache
- Dangling images
- Unused images
## What Remains Intact
- Running containers
- Persistent Volume Claims (PVCs)
- Kubernetes volumes
- Current pod states
- Cluster configuration
Execute the disk check first, show the results, then ask if the user wants to proceed with cleanup.

57
plugin.lock.json Normal file
View File

@@ -0,0 +1,57 @@
{
"$schema": "internal://schemas/plugin.lock.v1.json",
"pluginId": "gh:colek42/claude-plugins:nk-devops-tools",
"normalized": {
"repo": null,
"ref": "refs/tags/v20251128.0",
"commit": "3e587faa63f0c544b06de868b4d973e9d3f0778e",
"treeHash": "57573c94b8f55bef8991ac14d3b588057fc336ce43435edcba0987abc2246ee4",
"generatedAt": "2025-11-28T10:15:45.624208Z",
"toolVersion": "publish_plugins.py@0.2.0"
},
"origin": {
"remote": "git@github.com:zhongweili/42plugin-data.git",
"branch": "master",
"commit": "aa1497ed0949fd50e99e70d6324a29c5b34f9390",
"repoRoot": "/Users/zhongweili/projects/openmind/42plugin-data"
},
"manifest": {
"name": "nk-devops-tools",
"description": "Helm, ArgoCD, Vault best practices and disk space management for Minikube/Skaffold workflows",
"version": "1.0.0"
},
"content": {
"files": [
{
"path": "README.md",
"sha256": "4bce29b5f8f1e12184b22e89555a5f8edac7db9b5b2b19a57cf2e0fb7187e298"
},
{
"path": ".claude-plugin/plugin.json",
"sha256": "9c2884233fd851e307d263c5a48059599aaa0a638ae22951b8d25225ee020c67"
},
{
"path": "commands/disk-cleanup.md",
"sha256": "94ec42d30cc36f6c737d3eb01ed58a8b40141c17c7bd905e258885a97fd83db6"
},
{
"path": "skills/helm-practices/SKILL.md",
"sha256": "4a493c01a1b1389ff221d03b0bba4c8d519b8d8cf0317050a3b5e90e15adb8bf"
},
{
"path": "skills/argocd-management/SKILL.md",
"sha256": "0655da8f3d32d04d780c727af521e375cd94da16a49de2ee71ccd87b1f85dfe8"
},
{
"path": "skills/vault-secrets/SKILL.md",
"sha256": "36fd76b5c101da5ca3b7db79eebe0846977b61afa642f50d81e22f20f9049cb1"
}
],
"dirSha256": "57573c94b8f55bef8991ac14d3b588057fc336ce43435edcba0987abc2246ee4"
},
"security": {
"scannedAt": null,
"scannerVersion": null,
"flags": []
}
}

166
skills/argocd-management/SKILL.md Normal file
View File

@@ -0,0 +1,166 @@
# ArgoCD Management Best Practices
This skill provides ArgoCD application management patterns.
## When to Use
Activates when:
- Working with ArgoCD-managed applications
- Creating or modifying Kubernetes resources in ArgoCD repos
- Deploying applications via ArgoCD
## Core Principle
**Never manually create/modify Kubernetes resources when using ArgoCD.**
ArgoCD manages the lifecycle of your Kubernetes resources. Manual kubectl operations will be overwritten by ArgoCD's sync process.
## The ArgoCD Way
### Creating Resources
❌ **Wrong - Manual kubectl:**
```bash
kubectl apply -f deployment.yaml
kubectl apply -f service.yaml
```
✅ **Correct - ArgoCD Application CR:**
```bash
# Only kubectl apply for ArgoCD Application CRs
kubectl apply -f argocd-application.yaml
```
Then let ArgoCD handle the actual application resources.
### Application CR Pattern
```yaml
apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: my-app
namespace: argocd
spec:
project: default
source:
repoURL: https://github.com/org/repo
targetRevision: HEAD
path: k8s/manifests
destination:
server: https://kubernetes.default.svc
namespace: my-app
syncPolicy:
automated:
prune: true
selfHeal: true
```
## Workflow
### 1. Update Source Repository
```bash
# Modify your Kubernetes manifests or Helm charts in git
git add k8s/
git commit -m "update deployment configuration"
git push
```
### 2. Let ArgoCD Sync
```bash
# ArgoCD automatically detects changes and syncs
# Or manually trigger sync
argocd app sync my-app
# Check sync status
argocd app get my-app
```
### 3. Verify Deployment
```bash
# Use kubectl for read-only operations
kubectl get pods -n my-app
kubectl logs -n my-app deployment/my-app
kubectl describe deployment -n my-app my-app
```
## ArgoCD CLI Commands
### Application Management
```bash
# List applications
argocd app list
# Get application details
argocd app get my-app
# Sync application
argocd app sync my-app
# Check sync status
argocd app wait my-app --health
# View application logs
argocd app logs my-app
# Diff current state vs desired
argocd app diff my-app
```
### Debugging
```bash
# See why sync failed
argocd app get my-app
# View events
kubectl get events -n my-app
# Check ArgoCD controller logs
kubectl logs -n argocd deployment/argocd-application-controller
```
## Best Practices
✅ **Do:**
- Use `kubectl apply -f` ONLY for ArgoCD Application CRs
- Let ArgoCD handle all application resources
- Use git as the single source of truth
- Use automated sync policies for continuous deployment
- Use read-only kubectl commands for debugging
❌ **Don't:**
- Manually create/modify Kubernetes resources with kubectl
- Edit resources directly with `kubectl edit`
- Use `kubectl apply` for app resources in ArgoCD-managed namespaces
- Fight with ArgoCD by manually changing resources
## GitOps Workflow
1. **Code Change** → Commit to git repository
2. **ArgoCD Detects** → Automatic or manual sync trigger
3. **ArgoCD Applies** → Resources created/updated in cluster
4. **ArgoCD Monitors** → Health and sync status tracked
5. **Self-Heal** → Auto-corrects manual changes (if enabled)
## Emergency Override
If you absolutely must make a manual change:
```bash
# 1. Pause auto-sync temporarily
argocd app set my-app --sync-policy none
# 2. Make your manual change
kubectl apply -f emergency-fix.yaml
# 3. Update git to match your change
git add k8s/emergency-fix.yaml
git commit -m "emergency fix applied"
git push
# 4. Re-enable auto-sync
argocd app set my-app --sync-policy automated
```
**But prefer:** Make the change in git first, then let ArgoCD apply it.

110
skills/helm-practices/SKILL.md Normal file
View File

@@ -0,0 +1,110 @@
# Helm Best Practices
This skill provides Helm chart development and maintenance patterns.
## When to Use
Activates when:
- Working with Helm charts
- Modifying Chart.yaml dependencies
- Updating Helm values
- Deploying applications with Helm
## Dependency Management
### Critical Rules
**Always use `helm dependency update`** - NOT `helm dependency build`
```bash
# Correct
helm dependency update
# Wrong - Don't use this
helm dependency build
```
**After changing Chart.yaml dependencies:**
1. Run `helm dependency update` to fetch updated dependencies
2. Run `make check-deps` before committing changes
3. Pre-commit hooks automatically rebuild dependencies
### Common Workflow
```bash
# 1. Modify Chart.yaml
vim Chart.yaml
# 2. Update dependencies
helm dependency update
# 3. Check dependencies
make check-deps
# 4. Commit changes (pre-commit hook runs automatically)
git add Chart.yaml Chart.lock charts/
git commit -m "update helm dependencies"
```
## Helm Commands
### Testing Charts
```bash
# Template and check output
helm template <release-name> <chart-path>
# Template with specific values
helm template <release-name> <chart-path> -f values.yaml
# Dry-run install
helm install <release-name> <chart-path> --dry-run --debug
```
### Debugging
```bash
# Check current values
helm get values <release-name>
# Check all computed values
helm get values <release-name> --all
# Get manifest
helm get manifest <release-name>
```
## Best Practices
### Chart Structure
- Keep templates focused and modular
- Use helpers (`_helpers.tpl`) for repeated logic
- Document all values in `values.yaml` with comments
- Use semantic versioning for chart versions
### Values Organization
- Use nested structures for related values
- Provide sensible defaults
- Document required vs optional values
- Use `.Values` consistently
### Dependency Management
- Pin dependency versions explicitly
- Use repository URLs, not local paths in production
- Keep `Chart.lock` in version control
- Run `make check-deps` before committing
## Common Gotchas
❌ **Don't:**
- Use `helm dependency build` (use `update` instead)
- Forget to run `make check-deps` before committing
- Commit without running pre-commit hooks
- Hard-code values that should be configurable
✅ **Do:**
- Use `helm dependency update` for dependency changes
- Run `make check-deps` as part of your workflow
- Let pre-commit hooks rebuild dependencies automatically
- Make values configurable and well-documented

216
skills/vault-secrets/SKILL.md Normal file
View File

@@ -0,0 +1,216 @@
# Vault & Secrets Management
This skill provides HashiCorp Vault best practices for secrets management.
## When to Use
Activates when:
- Working with secrets in Kubernetes
- Managing Vault initialization and unsealing
- Using External Secrets Operator
- Setting up new environments
## Vault Lifecycle
### Initialization
**Command:** `make vault-init`
This initializes a new Vault instance and saves keys to `vault-init/` directory.
```bash
# Initialize Vault (first time only)
make vault-init
# Keys are saved to vault-init/ directory:
# - vault-init/unseal-key-1
# - vault-init/unseal-key-2
# - vault-init/unseal-key-3
# - vault-init/root-token
```
### Unsealing
**Command:** `make vault-unseal`
Required after pod restarts. Vault starts in a sealed state and must be unsealed with keys.
```bash
# Unseal Vault (after pod restart)
make vault-unseal
```
**Why unsealing is needed:**
- Vault pods restart (manual restart, node failure, updates)
- Vault starts sealed for security
- Must provide unseal keys to make Vault operational
## Security Best Practices
### Never Commit Vault Keys or Tokens
❌ **Don't:**
```bash
# Don't commit these files
git add vault-init/
git commit -m "vault keys" # NEVER DO THIS
```
✅ **Do:**
```bash
# Ensure vault-init/ is in .gitignore
echo "vault-init/" >> .gitignore
# Store keys securely offline
# - Password manager
# - Encrypted backup
# - Secure key management system
```
### Secrets in Code
❌ **Don't:**
```go
// Don't hardcode secrets
const apiKey = "sk-1234567890abcdef"
password := "supersecret123"
```
✅ **Do:**
```go
// Use environment variables from Vault
apiKey := os.Getenv("API_KEY")
// Or use External Secrets Operator
// which automatically creates k8s Secrets from Vault
```
## External Secrets Operator
**External Secrets Operator manages Kubernetes Secret creation from Vault.**
### How It Works
1. Secrets stored in Vault
2. ExternalSecret CR references Vault path
3. ESO fetches secret from Vault
4. ESO creates/updates Kubernetes Secret automatically
### Example ExternalSecret
```yaml
apiVersion: external-secrets.io/v1beta1
kind: ExternalSecret
metadata:
name: app-secrets
spec:
refreshInterval: 1h
secretStoreRef:
name: vault-backend
kind: SecretStore
target:
name: app-secrets
creationPolicy: Owner
data:
- secretKey: api-key
remoteRef:
key: secret/data/app
property: api_key
```
This creates a Kubernetes Secret named `app-secrets` with data from Vault.
## Common Workflows
### Storing a New Secret
```bash
# 1. Write secret to Vault
vault kv put secret/myapp/config \
api_key=abc123 \
db_password=secretpass
# 2. Create ExternalSecret CR
kubectl apply -f external-secret.yaml
# 3. Verify Kubernetes Secret was created
kubectl get secret app-secrets
kubectl describe secret app-secrets
```
### Rotating a Secret
```bash
# 1. Update secret in Vault
vault kv put secret/myapp/config api_key=newkey123
# 2. ESO automatically syncs (based on refreshInterval)
# Or manually trigger sync by deleting the k8s Secret
kubectl delete secret app-secrets
# 3. ESO recreates Secret with new value
kubectl get secret app-secrets -o yaml
```
### Debugging Secrets
```bash
# Check Vault status
kubectl exec -it vault-0 -- vault status
# Check if Vault is sealed
kubectl logs vault-0 | grep sealed
# Check External Secrets Operator
kubectl logs -n external-secrets deployment/external-secrets
# Check ExternalSecret status
kubectl describe externalsecret app-secrets
```
## Vault Initialization Checklist
When setting up Vault in a new environment:
- [ ] Run `make vault-init` to initialize Vault
- [ ] Securely save unseal keys and root token from `vault-init/`
- [ ] Add `vault-init/` to `.gitignore`
- [ ] Run `make vault-unseal` to unseal Vault
- [ ] Configure Vault policies and auth methods
- [ ] Set up External Secrets Operator integration
- [ ] Test secret retrieval and rotation
## Pod Restart Recovery
When Vault pods restart:
```bash
# 1. Check Vault status
kubectl get pods -l app=vault
# 2. Vault will be sealed - check logs
kubectl logs vault-0 | tail -20
# 3. Unseal Vault
make vault-unseal
# 4. Verify Vault is operational
kubectl exec vault-0 -- vault status
```
## Best Practices Summary
✅ **Do:**
- Initialize Vault with `make vault-init`
- Unseal Vault after restarts with `make vault-unseal`
- Use External Secrets Operator for Kubernetes secrets
- Store unseal keys securely offline
- Never commit Vault keys to git
- Rotate secrets regularly
❌ **Don't:**
- Commit vault-init/ directory to git
- Hardcode secrets in code
- Manually create Kubernetes Secrets for Vault-backed data
- Leave Vault sealed
- Share root token unnecessarily