Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From docker-maintenance
Update Docker base images with sha256 pinning, manage Containerfile dependencies, test builds, and keep an audit log. Use when updating containers or checking for stale images.
npx claudepluginhub cameri/claude-skills --plugin docker-maintenanceHow this skill is triggered — by the user, by Claude, or both
Slash command
/docker-maintenance:docker-maintenanceThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
<essential_principles>
Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Explores codebases via GitNexus: discover repos, query execution flows, trace processes, inspect symbol callers/callees, and review architecture.
Share bugs, ideas, or general feedback.
<essential_principles> <one_at_a_time> Always update ONE container per invocation. If no target is specified, pick the one with the oldest entry in the update log (or no entry at all). Never batch updates — each must be tested before proceeding to the next. </one_at_a_time>
<containers_root>
All services live under /workspace/containers/. Each service has its own directory with a compose.yml. Some also have a Containerfile or Dockerfile for custom images.
</containers_root>
<update_log>
The update log lives at /workspace/containers/UPDATE-LOG.md. Every update attempt — successful or reverted — gets a log entry. This is the source of truth for "oldest updated."
</update_log>
<safety_first> Before any change: capture the current state (image digest, file content). If the update fails testing, revert immediately and log the failure. Never leave a service broken. </safety_first>
<sha256_pinning>
For compose images, prefer pinning with sha256 digest alongside the tag:
image: nginx:1.27.3@sha256:abc123...
This ensures reproducibility even if the tag is moved.
</sha256_pinning>
<version_control>
Use jj (not git) for all version control operations. See /workspace/CLAUDE.md for jj commands.
</version_control>
</essential_principles>
Or just describe what you want and I'll route appropriately.
| Response | Workflow | |----------|----------| | 1, "update", "maintain", container name given | `workflows/update-container.md` | | 2, "health", "check", "status", "unhealthy" | `workflows/check-health.md` | | 3, "add", "new service", "onboard" | `workflows/add-service.md` | | 4, "remove", "delete", "uninstall" | `workflows/remove-service.md` | | 5, "gatus", "monitor", "alert", "endpoint" | `workflows/manage-gatus.md` | | 6, "log", "history", "view" | `workflows/view-log.md` | | 7, "init", "initialize", "create log" | `workflows/view-log.md` (init section) |After reading the workflow, follow it exactly.
<reference_index> Container types and how to handle them: references/container-types.md Update strategies and sha256 pinning: references/update-strategies.md Log format and entry structure: references/log-format.md Infrastructure specifics (networks, deps, Watchtower, Gatus): references/environment.md </reference_index>
<workflows_index>
| Workflow | Purpose |
|---|---|
| update-container.md | Full update cycle: pick target → update → test → log |
| check-health.md | Survey unhealthy containers and diagnose failures |
| add-service.md | Onboard a new service: compose, healthcheck, Gatus, TsDproxy |
| remove-service.md | Clean removal: compose, Gatus, TsDproxy, volumes |
| manage-gatus.md | Add/remove/modify Gatus endpoints and alert providers |
| view-log.md | Show update history; initialize log if missing |
| </workflows_index> |