A collection of production-ready Helm charts for deploying applications to Kubernetes.
- drunk-app — main application chart
- doc: README.md · Full docs
- repo url: https://baoduy.github.io/drunk.charts/drunk-app
- drunk-lib — reusable library chart (dependency of drunk-app)
- doc: README.md
- repo url: https://baoduy.github.io/drunk.charts/drunk-lib
- drunk-nginx-gateway — Nginx gateway chart
- drunk-traefik-gateway — Traefik gateway chart
- drunk-squid-basic-auth — Squid proxy with basic auth
helm repo add drunk-charts https://baoduy.github.io/drunk.charts/drunk-app
helm repo update
helm install my-app drunk-charts/drunk-app -f my-values.yamlAn AI assistant plugin is available to help configure values.yaml for drunk-app inside Claude Code.
Install the plugin:
claude plugin marketplace add baoduy/drunk.charts --scope project
claude plugin install drunk-app --scope projectInstall the SKILL to your local project:
After installing the plugin, copy the SKILL.md file into your project so Claude Code can discover it locally:
mkdir -p .claude/skills
cp "$(plugin path drunk-app)/skills/drunk-app/SKILL.md" .claude/skills/drunk-app.mdAlternatively, you can manually download the skill file:
mkdir -p .claude/skills
curl -sL https://raw.githubusercontent.com/baoduy/drunk.charts/main/plugins/drunk-app/skills/drunk-app/SKILL.md -o .claude/skills/drunk-app.mdNote: The
.claude/skills/directory must exist in your project root for Claude Code to detect the skill. Commit this file to your repository so all team members benefit from the AI assistant.
Then use /drunk-app in any Claude Code session to get contextual help with chart configuration.
Charts are tested with helm-unittest.
Install the plugin:
helm plugin install https://github.com/helm-unittest/helm-unittest.gitRun tests:
helm unittest -f 'tests/*.yaml' ./drunk-app
helm unittest -f 'tests/*.yaml' ./drunk-libOr use the provided test scripts:
./drunk-app/test.sh
./drunk-lib/verify.sh