Verified:Website 𝕏@q_yeon_gyu_kim GitHub

mass-ulw

Orchestrate multi-agent dependency graphs for complex task execution and staged workflow automation.

68.6k|5.6k|Updated Dec 3, 2025

Search Tags:#Automation #Workflow #dependency-graph #DAG # orchestration #Multi-Agent

One-click install

npx skills add https://github.com/code-yeongyu/oh-my-opencode --skill mass-ulw

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: mass-ulw
Source: https://github.com/code-yeongyu/oh-my-opencode/tree/main/packages/omo-senpi/skills/mass-ulw
Command: npx skills add https://github.com/code-yeongyu/oh-my-opencode --skill mass-ulw

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill addresses the challenge of managing complex, multi-step projects where tasks have strict ordering dependencies, preventing the inefficiency of manual coordination or the failure of unmanaged parallel execution.

Core Features & Use Cases

Dependency Graph Orchestration: Define and execute directed acyclic graphs (DAGs) where tasks fan-out and fan-in based on completion status.
Staged Workflow Control: Manage multi-agent jobs with built-in support for retries, node-level steering, and adaptive graph amendments.
Use Case: Automate a multi-stage documentation refresh by auditing stale API references, rewriting affected pages, and verifying internal links in a single orchestrated run.

Quick Start

Use the mass-ulw skill to define and start a dependency graph for your multi-agent task.

Frequently Asked Questions about mass-ulw

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

FAQPage Schema

How do I define a dependency graph for multi-agent tasks?▼

Use the sdk.define method to create a DAG, then add nodes with specific prompts, categories, and dependsOn identifiers to establish the execution order.

What is the difference between a task spawn and a mass-ulw workflow?▼

Task spawns are for independent parallel work, while mass-ulw workflows are for complex jobs where task ordering and dependency management are critical to success.

Can I recover a failed node without restarting the entire graph?▼

Yes, use the sdk.retry method to re-run only the failed or cancelled nodes, allowing completed nodes to retain their cached results.

When should I use the deep category for a node?▼

Reserve the deep category for complex debugging or cross-module reasoning tasks that cannot be handled by standard quick or unspecified-low categories.

How do I handle data flow between dependent nodes?▼

Dependency edges in mass-ulw are for ordering only; ensure each node prompt is self-contained and does not rely on upstream output substitution.