docker-manager

Plugin docker-manager in danielrosehill/docker-manager-plugin. Commands read from the repo's manifest — 9 days ago.

Skills (12)

• adopt-deployment

  Remediate a Docker stack that was brought up loosely - no git repo, compose file in a random place like /root or /tmp, or a container started by hand with docker run - into the intended shape - its own repo (asks public or private), secrets and data gitignored, living under the deployment root, adde

• audit-autostart

  Audit what starts automatically on a Docker host - whether the Docker daemon is enabled at boot, each container's restart policy (no, always, unless-stopped, on-failure), systemd units or @reboot cron jobs that start containers outside Docker, and containers that would not come back after a reboot -

• audit-deployments

  Check where each running Docker stack is deployed from - every container traced to its compose project, working directory and compose file, whether that sits under the user's deployment root, is a git repo with a remote and no uncommitted drift, and flag containers started outside compose, images bu

• audit-engine

  Check whether the Docker Engine, containerd and the Compose plugin are up to date on a host - installed vs available package versions, install method (apt, snap, static, Docker Desktop), whether the apt repo matches the OS release, and daemon settings worth knowing (live-restore, log driver, log rot

• audit-storage

  Audit container storage on a Docker host - which running containers have persistent storage and which don't (state that would be lost on recreate), where each bind mount and named volume actually lives on disk, which sit outside the user's common volume root, anonymous volumes, orphaned volumes, and

• audit-watchtower

  Check Watchtower on a Docker host - is it installed and running, which fork and version, is the image current, its schedule and cleanup settings, which containers it updates or skips, and its update history (what it updated, when, failures, the last completed session). Read-only. Use for "is watchto

• deployment-master

  Set up and run the master deployment repo shape - one git repo per Docker stack, all added as submodules of a single master repo checked out at the host's deployment root, with a stacks.yaml manifest (tier, enabled) and bin/up.sh and bin/status.sh to bring everything up and check drift. Use for "cre

• engine-auto-update

  Keep Docker Engine, containerd and the Compose plugin updated automatically on Ubuntu/Debian hosts by adding Docker's apt repo to unattended-upgrades, then prove it with a dry run - including the traps that silently block it (wrong origin pattern, held packages, stale package lists aborting the whol

• fleet-audit

  Full read-only audit of one or all Docker hosts in the user's preferences - host load and memory, containers running/unhealthy/restarting, engine version and pending updates, Watchtower status and history, autostart and restart policies, persistent storage and volume locations, deployment sources, a

• preferences

  Record or change how the user's Docker hosts are meant to be set up (hosts to manage and how to reach them, where persistent volumes should live, where deployments come from, autostart and restart-policy rules, Watchtower expectations, cleanup limits) in a preferences file outside the plugin. Use fo

• swap-relief

  Diagnose and clear a Docker host's full swap - which processes and containers hold swapped memory, whether RAM can absorb it, then a safe swapoff/swapon cycle, plus the lasting fixes (container memory limits, swappiness, more RAM or swap). Use for "swap is full", "clean the swap", "server is swappin

• watchtower-remediate

  Fix the problems audit-watchtower finds - an image that no longer resolves (registry 404 every run, e.g. an upstream that stopped publishing), the unmaintained containrrr/watchtower that breaks on new Docker APIs, containers that must not auto-update (databases), missing or broken schedules, stale s

Slash commands (7)

• /docker-manager:backup

  Backup Docker volumes and container data to a local directory or remote storage.

• /docker-manager:cleanup

  Analyze unused Docker resources and guide the user through safe cleanup options.

• /docker-manager:deploy

  Deploy a new Docker Compose stack or manage an existing one.

• /docker-manager:logs

  View and analyze logs for a Docker container.

• /docker-manager:setup

  Probe the local Docker environment and generate initial context files for the workspace.

• /docker-manager:status

  Perform a comprehensive health check of all Docker containers and resources.

• /docker-manager:update-context

  Refresh workspace context files with the current Docker environment state.

Subagents (5)

• compose-generator

  Generate production-ready Docker Compose files for common services and multi-service stacks.

• container-diagnostics

  Diagnose issues with Docker containers including crashes, restarts, and connectivity problems.

• migration-planner

  Plan and execute Docker container migrations between environments with minimal downtime.

• resource-optimizer

  Analyze Docker resource usage (CPU, memory, storage) and produce actionable optimization recommendations.

• security-audit

  Audit a Docker environment for security misconfigurations and produce a prioritised remediation report.