docker-ops

name	docker-ops
description	Docker compose runbook for the 3D Printer Slicer API. Use this when asked to start, stop, restart, monitor, or clean up the containerized backend environment.

Use this skill for container lifecycle operations in this repository.

Slash entrypoint:

• Use /docker-ops for startup, restart, logs, health checks, and cleanup operations.

Full agent definitions with architecture context, security configuration, hard rules, and scope boundaries are mirrored in .github/agents/docker-specialist.md and .claude/agents/docker-specialist.md. Read that file for complete context before performing Docker operations.

Quick Command Reference

1. Standard backend

   • Start: docker compose up -d --build
   • Stop: docker compose down
   • Logs: docker compose logs -f slicer-api
   • Health check: curl http://127.0.0.1:3000/health

2. Development mode (live mount)

   • Start: docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d --build
   • Stop: docker compose -f docker-compose.yml -f docker-compose.dev.yml down

3. Monitoring mode

   • Start: docker compose --profile monitoring up -d --build
   • Stop monitoring only: docker compose stop uptime-kuma

4. Safe partial cleanup (preferred)

   • docker compose down -v --remove-orphans
   • docker image prune -f
   • docker network prune -f

5. Global cleanup (destructive, explicit user approval required)

   • docker compose down -v --remove-orphans
   • docker system prune -a --volumes -f

Execution Workflow

1. Identify requested operation scope (standard, dev, monitoring, or cleanup).
2. Run the minimal command set needed for that scope.
3. Verify health with GET /health and targeted logs.
4. Report service status and any follow-up actions.

Safety Rules

• Never run docker system prune -a unless user explicitly requests destructive global cleanup.
• If EACCES errors appear on mounted files, review SLICER_CONTAINER_USER in .env and rebuild.
• If monitoring UI is unreachable, verify startup used --profile monitoring.

Validation Checklist

• Requested containers are up (or cleanly stopped) with expected profiles.
• Health endpoint and logs reflect expected state.
• No destructive cleanup executed without explicit approval.