Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From mega-code
Stops a running MEGA-Code skill extraction pipeline by run ID or interactive selection. Useful for aborting stuck or unnecessary pipeline processes.
npx claudepluginhub wisdomgraph/mega-code --plugin mega-codeHow this skill is triggered — by the user, by Claude, or both
Slash command
/mega-code:stop [run-id][run-id]This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Stop a currently running MEGA-Code skill extraction pipeline.
Stops running Output SDK workflows via `npx output workflow stop`. Use to cancel unnecessary, stuck, long-running, or erroneous executions.
Gracefully stops active autonomous workflows, preserves progress, summarizes status, and generates minimal recovery plans with handoffs. Useful for safe pauses on user cancel requests.
Share bugs, ideas, or general feedback.
Stop a currently running MEGA-Code skill extraction pipeline.
MEGA_DIR="$(cd "${CLAUDE_SKILL_DIR}/../.." && pwd)"
export MEGA_CODE_AGENT="claude"
uv run --directory "$MEGA_DIR" python -m mega_code.client.check_auth
If the auth check fails (non-zero exit), show the output to the user and stop.
Skip to the Stop step below using the provided run-id.
Step 1 — List active runs:
uv run --directory "$MEGA_DIR" mega-code pipeline-status
If the output says "No active pipeline runs.", tell the user and stop.
Step 2 — Ask user to confirm:
Use the AskUserQuestion tool to present the active runs and let the user choose.
Always include a cancel option — even if there is only one active run.
Format the question like:
Active pipeline runs:
1. {run_id} | project: {project_id} | status: {status}
Phase: {current_phase} ({sessions_processed}/{sessions_total})
Which run would you like to stop? Select a number, or 0 to cancel.
If user selects 0 or cancels, say "Cancelled." and stop.
uv run --directory "$MEGA_DIR" mega-code pipeline-stop --run-id <SELECTED_RUN_ID>
Report the result to the user. If successful, confirm: "Pipeline {run_id} has been stopped."