agentleFS
WhispersOfJWhispersOfJ/arrfleet-toolkit/skills/docker-compose-manager/SKILL.md

docker-compose-manager

Skill1 starsChanged 4 months agoMITView on GitHub

Manage an Arr-stack docker compose project safely — start, stop, restart, and inspect containers. Use whenever the user asks to restart a service, bring the stack up/down, check container status, tail logs, or recover from a container crash. Supports an optional mount-cascade map for FUSE/network-mount-backed setups (e.g. rclone, NzbDAV-rclone) so a mount owner's dependents restart in the correct order instead of being left pointed at a stale mount. Trigger phrases: "restart <service>", "bring the stack up/down", "is <container> healthy", "recreate <service>", "docker compose logs for X", "the mount looks stale".

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    mount-owning containers cascade: If a service in this stack owns a FUSE or network-remount mount (rclone, NzbDAV-rclone, etc.) that other …

    line 9
  2. Very specific:

    Run from the repo root (where docker-compose.yml lives), or pass --compose-dir /path/to/repo.

    line 41
  3. Very specific:

    Never pass --force flags or skip the confirmation prompt on down even if asked to "just do it" — surface the prompt and let the user …

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Docker Compose Manager

Wraps docker compose for an Arr-stack project with one optional safety behavior: mount-owning containers cascade. If a service in this stack owns a FUSE or network-remount mount (rclone, NzbDAV-rclone, etc.) that other containers bind-mount, a plain restart of the owner alone leaves every dependent holding a stale mount handle until they're restarted too — a common failure class in these stacks. A plain restart also reuses the owner's existing mount namespace and never picks up a fresh mount at all, so a cascade restart force-recreates the owner and every dependent instead of restarting them.

When to use this skill

  • Restarting any single service or group of services
  • Bringing the whole stack up or down
  • Checking docker compose ps / health status across all containers
  • Tailing logs for a specific container
  • Recreating a container after an image or config change (--force-recreate) — required after editing a single-file bind mount, since up -d alone can silently keep serving the stale pre-edit file to a running container

Usage

python3 handler.py status                       # docker compose ps, all services
python3 handler.py status radarr                 # status of one service
python3 handler.py restart radarr                # restart one non-mount-owning service
python3 handler.py restart nzbdav-rclone         # force-recreate it AND its cascade dependents, in order (if configured)
python3 handler.py up                            # docker compose up -d (whole stack)
python3 handler.py down                          # docker compose down (whole stack) — asks for confirmation
python3 handler.py recreate radarr               # force-recreate a single container (config/mount changed)
python3 handler.py logs radarr --tail 200
python3 handler.py cascade-map                   # print the known mount-owner -> dependent graph

Run from the repo root (where docker-compose.yml lives), or pass --compose-dir /path/to/repo.

Configuring the cascade map

By default no service is treated as a mount owner. If this stack has one (a service whose restart leaves other containers holding a stale mount), define the topology in .arrfleet-cascade.json next to docker-compose.yml:

{"nzbdav-rclone": ["radarr", "sonarr", "plex", "unpackerr"]}

This can't be inferred from docker-compose.yml alone — a bind mount doesn't declare "this is a remount point." Update the file whenever the compose topology changes.

Safety rules

  • down always prompts for confirmation before executing — it stops the entire stack.
  • restart/recreate on a single service never touches unrelated services unless that service is a known mount owner in .arrfleet-cascade.json.
  • Never pass --force flags or skip the confirmation prompt on down even if asked to "just do it" — surface the prompt and let the user confirm explicitly, since this is a shared, hard-to-reverse action against a running media server other people may be using.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.