ultrawork

Coordinate parallel task execution with structured state and evidence lanes.

Updated Jan 4, 2026
One-click install
npx skills add https://github.com/YeaChan05/Remittance --skill ultrawork-yeachan05
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ultrawork
Source: https://github.com/YeaChan05/Remittance/tree/main/.codex/skills/ultrawork
Command: npx skills add https://github.com/YeaChan05/Remittance --skill ultrawork-yeachan05

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ultrawork coordinates parallel task execution to accelerate completion while maintaining disciplined context and lightweight evidence, without providing persistence guarantees.

Core Features & Use Cases

  • Gather context before execution and define clear pass/fail acceptance criteria up front.
  • Run parallel execution lanes along with optional evidence lanes to validate progress concurrently.
  • Use structured state management and explicit stop/escalation conditions to keep workflows safe and auditable.

Quick Start

In a scenario with multiple independent tasks, activate ultrawork to run them in parallel while generating lightweight evidence for session verification.

Frequently Asked Questions about ultrawork

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

FAQPage Schema
How do I run parallel execution for multiple independent tasks concurrently?

Parallel execution coordinates multiple concurrent task lanes to maximize throughput, applying direct-tool execution with optional background evidence lanes for lightweight session verification.

What is task orchestration and how does it manage concurrent workflow state?

Task orchestration uses structured state management and explicit stop or escalation conditions to keep concurrent workflows safe, auditable, and aligned with upfront acceptance criteria.

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

Use parallel execution lanes when multiple independent tasks can run concurrently to accelerate completion, generating lightweight evidence without persistence guarantees.

How do I define pass/fail acceptance criteria before running concurrent tasks?

Gather context before execution and define clear pass/fail acceptance criteria upfront, then coordinate parallel execution lanes along with optional evidence lanes to validate progress concurrently.

What are the limitations of lightweight evidence in parallel task orchestration?

Lightweight evidence in parallel task orchestration provides session verification without persistence guarantees, meaning generated evidence is not permanently stored beyond the active workflow context.

Can I use ultrawork to maximize throughput without persistence guarantees?

Ultrawork coordinates parallel task execution to maximize throughput with lightweight evidence while explicitly not providing persistence guarantees for long-term state storage.