ultrawork

Coordinate parallel task execution across multiple agents with tier-aware routing.

1|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill ultrawork-oneforce
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ultrawork
Source: https://github.com/oneforce/oh-my-gemini-cli/tree/main/skills/ultrawork
Command: npx skills add https://github.com/oneforce/oh-my-gemini-cli --skill ultrawork-oneforce

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Independent work often stalls when tasks are executed sequentially. Ultrawork enables firing multiple agents concurrently to reduce wait times and improve throughput.

Core Features & Use Cases

  • Parallel agent execution across multiple tasks to accelerate completion.
  • Deterministic routing to the appropriate agent tier (haiku/sonnet/opus) based on task complexity.
  • Optional background execution for long-running operations to keep overall workflows responsive.
  • Use cases include orchestrating concurrent validation, data processing, or exploratory tasks within a larger lifecycle.

Quick Start

Provide a parallel task execution plan by firing independent tasks to the correct agent tiers with optional background execution.

Frequently Asked Questions about ultrawork

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

FAQPage Schema
How do I execute independent tasks in parallel across multiple agents?

Parallel task execution fires multiple agents concurrently to reduce wait times. It applies deterministic task dispatch to accelerate independent work and improve overall workflow throughput.

What is tier-aware routing for multi-agent orchestration?

Tier-aware routing is a deterministic dispatch mechanism that assigns tasks to appropriate agent tiers (haiku, sonnet, opus) based on complexity. It ensures each parallel task matches the correct processing tier.

How do I handle long-running operations during parallel execution without blocking the workflow?

Background execution keeps workflows responsive by running long operations separately. It allows the main multi-agent orchestration to continue processing other independent tasks without stalling.

When should I use parallel execution instead of sequential task processing?

Parallel execution is needed when independent tasks stall during sequential processing. Use it for concurrent validation, data processing, or exploratory tasks within a larger lifecycle to maximize throughput.

Does multi-agent orchestration work for dependent tasks that require routing to specific agent tiers?

Multi-agent orchestration applies to independent tasks requiring routing to the correct agent tier. It implements deterministic task dispatch and in-context guidance for agent selection, excluding tightly coupled dependent workflows.