One-click install
npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ultrawork-marcmunoz-uno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ultrawork
Source: https://github.com/marcmunoz-uno/jailbreak/tree/main/skills/ultrawork
Command: npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ultrawork-marcmunoz-uno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ultrawork eliminates wasted time from running independent subtasks sequentially by executing them concurrently.

Core Features & Use Cases

  • Parallel agent execution: Launch multiple agent calls at once for tasks that don’t depend on each other.
  • Smart model routing: Select appropriate model tiers (LOW/MEDIUM/HIGH) based on task complexity to reduce cost while maintaining quality.
  • Composable execution component: Designed as a parallelism layer that does not provide persistence, verification loops, or state management by itself.

Quick Start

Ask your AI to run ultrawork to dispatch three independent subtasks (e.g., implement features, add tests, update documentation) concurrently using appropriate model tiers.

Frequently Asked Questions about ultrawork

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

FAQPage Schema
How do I run independent agent tasks in parallel to speed up workflow execution?

To run independent agent tasks in parallel, you dispatch multiple concurrent agent calls for subtasks that do not depend on each other. This accelerates workflow execution by eliminating the wasted time of sequential processing.

What is model-tier routing for concurrent tasks and how does it reduce costs?

Model-tier routing assigns appropriate model complexity (LOW/MEDIUM/HIGH) to concurrent tasks based on their specific requirements. This reduces overall execution costs by using lightweight models for simple tasks while maintaining quality for complex ones.

How do I execute long-running background jobs without blocking dependent work?

You can execute long-running operations as background jobs to prevent blocking dependent work. This satisfies execution policy requirements by allowing concurrent firing for independent tasks while long-running operations process asynchronously.

Does parallel agent execution provide state management and verification loops?

No, parallel agent execution does not provide state management or verification loops. It is designed strictly as a composable parallelism layer, meaning you must integrate external tools if your workflow requires persistence or automated verification.

When should I not use concurrent tasks for agent orchestration?

You should not use concurrent tasks for agent orchestration when subtasks are dependent on each other. This approach requires tasks to be classified as parallel-safe, meaning any workflow with sequential dependencies will not benefit from this concurrent execution.