メニュー
Lists all examiner tasks (init and regular) for a playground. Use to check the status of challenge/tutorial tasks.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Debugs an existing challenge by starting it, inspecting tasks, troubleshooting over SSH, fixing source files, and restarting. Use when a challenge's tasks are failing or hanging.
Gets detailed status and output of a specific examiner task on a playground. Use to debug why a task is failing.
Executes a command on a playground VM via SSH. Use when you need to run commands on a playground for testing, debugging, or verification.
Applies a scripted solution to a running challenge (single- or multi-VM) and verifies every non-helper task completes. Use to validate a .solution.sh / .solution-NN.sh end-to-end.
Starts a challenge playground session. Use when you need to launch a challenge for testing or solving.
Stops the current solution attempt for a challenge. Use when you need to stop a running challenge.
| name | list-playground-tasks |
| description | Lists all examiner tasks (init and regular) for a playground. Use to check the status of challenge/tutorial tasks. |
| argument-hint | <playground-id> |
Default (preferred): list tasks via the API. This works for single- and multi-machine playgrounds, shows tasks from all machines, and needs no SSH tunnel - so it's the right tool for polling in a loop while waiting on init or verification:
labctl playground tasks $0 -o yaml
Because this path is tunnel-free, prefer it over SSH for status polling - it
avoids exhausting the SSH tunnel slot pool (see the SSH tunnel discipline notes in
the run-playground-command skill).
Tasks are edge-activated (not level-triggered) and their status never regresses:
the examiner keeps evaluating a task until it first observes a success or failure
state, then it settles into its terminal state - completed (40) or failed
(35) - and never runs again, even if the world state later changes. The numerical
status only ever increases. So when polling, stop as soon as every non-helper task
has reached 40; a passed task cannot later flip back or fail.
The SSH-based alternative lists tasks for a single machine only (the one you SSH into) and is rarely needed now:
labctl ssh $0 -- examinerctl task ls
# or, for a specific machine:
labctl ssh $0 --machine <machine-name> -- examinerctl task ls
Use the get-playground-task skill with a specific task name to see detailed output
including stdout/stderr of the last execution.