cancel

Cancel active OMC workflows and stop linked components in dependency order.

3|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/marcmunoz-uno/jailbreak --skill cancel-marcmunoz-uno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cancel
Source: https://github.com/marcmunoz-uno/jailbreak/tree/main/skills/cancel
Command: npx skills add https://github.com/marcmunoz-uno/jailbreak --skill cancel-marcmunoz-uno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Canceling an active autonomous workflow can leave sessions stuck, state files inconsistent, or linked modes (like persistence loops and QA workers) still running. This Skill resolves that by detecting the currently active OMC mode(s) and stopping the right components in the correct dependency order while preserving any required resume state.

Core Features & Use Cases

  • Autodetects active OMC mode(s) across session-scoped workflows (autopilot, ralph, ultrawork, ultraqa, swarm, ultrapilot, pipeline, team, plan consensus).
  • Dependency- and link-aware cancellation ensures correct teardown ordering (e.g., autopilot cascades to linked ralph/ultraqa cleanup; ralph cascades to linked ultrawork).
  • Force/global cleanup support clears all sessions plus legacy artifacts when you truly need a full reset.
  • Team graceful shutdown for Claude Code teams signals teammates, waits for shutdown responses, then performs registry and state cleanup with an orphan-process scan.
  • Emergency fallback provides a last-resort state-file cleanup path when state tools are unavailable (with explicit guardrails for modes that require special handling).

Quick Start

Use the cancel command to stop the currently running OMC mode: set off the flow with /oh-my-claudecode:cancel.

Frequently Asked Questions about cancel

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I cancel an autonomous agent workflow without leaving stuck processes?

Canceling an autonomous workflow safely requires detecting the active mode and stopping linked components in the correct dependency order. This prevents stuck sessions and inconsistent state files while preserving required resume state.

How does dependency-aware workflow cancellation handle linked modes like autopilot and ralph?

Dependency-aware workflow cancellation ensures correct teardown ordering by cascading stops across linked modes. Canceling autopilot triggers linked ralph or ultraqa cleanup, and ralph cascades to linked ultrawork, preventing orphaned processes.

Can I force a global cleanup to clear all active sessions and legacy artifacts?

Yes, force global cleanup clears all active sessions and legacy artifacts when you need a full reset. This clears session-scoped autonomous workflows like swarm, ultrapilot, pipeline, and team modes across the environment.

What is the emergency fallback for state cleanup when state management tools fail?

The emergency fallback for state cleanup is a last-resort file-removal path used when state tools are unavailable. It includes explicit guardrails for modes requiring special handling to avoid corrupting persistent state files during recovery.

How do I perform a graceful team shutdown for Claude Code teams?

A graceful team shutdown for Claude Code teams signals teammates, waits for shutdown responses, then performs registry and state cleanup. This is followed by an orphan-process scan to ensure no background workers remain active.

Why does canceling an active workflow cause inconsistent state files and debugging recovery issues?

Canceling an active workflow causes inconsistent state files when linked modes like persistence loops and QA workers keep running after the main process stops. Proper session-aware state clearing and dependency-ordered teardown prevent these debugging recovery issues.