Docker containers
WECOMMANDER understands that a Linux server can run more than one Docker daemon: the system one and a rootless one per user. Every Docker feature names the daemon it works on.
In the desktop app
- Disk shows Docker's usage (images, containers, volumes) and marks the storage folder of each daemon.
- Cleanup frees each daemon's build cache and unused images through Docker's own prune; running containers and volumes are never touched.
- Processes shows which container owns a process, and can restart that container to clear its zombies.
For AI
| Tool | What it does | Approval |
|---|---|---|
| docker_containers_list | Containers per daemon (system, or u<uid> for a rootless one): id, name, image, state, status, created, ports and restart count. A daemon that cannot be read shows its error. | Read-only |
| docker_container_logs | The last lines of one container's logs (stdout and stderr, timestamped). tail 1–2000 (default 200), since such as 10m or 1h30m, at most 256 KiB; secrets are masked. | Read-only |
| docker_container_action | start, stop or restart one container (stop and restart wait 10 seconds before killing). | Every call needs an approval unless the device or SSH server is on full trust; never allowlisted; refused on read-only SSH servers |
Guarantees
- Containers, images and volumes are never removed or pruned by these tools.
- WECOMMANDER's own containers are never stopped or restarted.
- Container references are checked strictly before anything reaches the server, and the action runs on the full id Docker resolves.
- Available on Linux agent devices and on SSH servers shared through Hybrid Access.
- Disk usage and safe cleanup (including rootless Docker): Find what fills a Linux server's disk and free space safely: old logs, package caches, crash dumps, and Docker build cache and unused images per Docker daemon, including rootless Docker.
- MCP tool reference: Every WECOMMANDER MCP tool, grouped: what it does, its risk and approval rule, where it works (agent devices, SSH servers, Linux only) and its key arguments.
- AI permissions: trust levels, allowed commands, root and the safety net: Set per device and per SSH server what AI may do without asking: strict, standard or full trust (time-limited or not), allowed command patterns, run as root with two keys, the safety net and autonomous mode.