task-decomposition

Decompose complex software tasks into atomic subtasks structured as a DAG.

1|Updated May 21, 2026
One-click install
npx skills add https://github.com/hiddink-ai/hiddink-harness --skill task-decomposition-hiddink-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: task-decomposition
Source: https://github.com/hiddink-ai/hiddink-harness/tree/main/templates/skills/task-decomposition
Command: npx skills add https://github.com/hiddink-ai/hiddink-harness --skill task-decomposition-hiddink-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of managing large, multi-faceted coding tasks that are too complex for a single agent to handle effectively, preventing context overload and ensuring efficient execution.

Core Features & Use Cases

  • Automated Decomposition: Breaks down high-level requirements into atomic, parallelizable subtasks based on domain, file count, and agent expertise.
  • DAG Orchestration: Generates structured workflow specifications that define dependencies and execution order.
  • Granularity Guardrails: Automatically validates subtask size against file limits to ensure optimal agent performance and prevent execution errors.
  • Use Case: When tasked with implementing a full-stack feature, the skill automatically splits the work into separate database, API, frontend, and testing subtasks, assigning each to the appropriate specialist agent.

Quick Start

Use the task-decomposition skill to analyze the current project requirements and generate a parallel execution plan for the upcoming feature implementation.

Frequently Asked Questions about task-decomposition

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

FAQPage Schema
How do I break down complex software development tasks for parallel execution?

Breaking down complex software development tasks involves decomposing high-level requirements into atomic, parallelizable subtasks structured as a Directed Acyclic Graph (DAG). This prevents context overload by mapping subtasks to specialized agents based on domain and file count.

What is a DAG workflow for multi-agent coding tasks?

A DAG workflow for multi-agent coding is a Directed Acyclic Graph that structures automated decomposition by defining subtask dependencies and execution order. It ensures large coding tasks are split into granular assignments for specialized agents.

How do I split a full-stack feature implementation into specialized agent subtasks?

To split a full-stack feature implementation, automated decomposition analyzes project requirements and divides the work into separate database, API, frontend, and testing subtasks. Each subtask is assigned to the appropriate specialist agent for parallel execution.

Do I need orchestration and routing skills to manage task decomposition?

Yes, task decomposition requires integration with dag-orchestration and routing skills to manage multi-agent workflows. These dependencies ensure generated subtask specifications execute correctly and maintain task atomicity across the orchestrator agent context.

What are the granularity limits for subtasks in a parallel execution plan?

Granularity limits for subtasks are enforced by validating subtask size against file-count limits. This ensures optimal agent performance and prevents execution errors by keeping automated decomposition tasks atomic and manageable for specialized agents.