parallel-decompose

Decomposes PRDs and Tech Specs into parallel-executable tasks with contracts and dependency graphs.

14|1|Updated Dec 5, 2025
One-click install
npx skills add https://github.com/jpoutrin/product-forge --skill parallel-decompose
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parallel-decompose
Source: https://github.com/jpoutrin/product-forge/tree/main/plugins/product-design/skills/parallel-decompose
Command: npx skills add https://github.com/jpoutrin/product-forge --skill parallel-decompose

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill guides decomposing PRDs into parallel-executable tasks, defining contracts, waves, and task artifacts for cpo execution.

Core Features & Use Cases

  • PRD decomposition workflow from TS/PRD into manifest.json, context.md, and tasks
  • Tech Spec analysis and contract generation
  • Generates prompts and task graphs for cpo

Quick Start

Use the workflow to transform a PRD into a parallel directory ready for cpo.

Frequently Asked Questions about parallel-decompose

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

FAQPage Schema
How do I decompose a PRD into parallel tasks?

Decompose PRDs into parallel-executable tasks by analyzing your PRD or Tech Spec, generating a manifest.json with task definitions, contracts (types.py, api-schema.yaml), and a task graph. The Skill produces context.md, architecture.md with Mermaid diagrams, and task-specific prompts ready for multi-agent execution.

What are task contracts and how do I generate them?

Task contracts define interfaces and expectations for parallel work. This Skill generates contracts in types.py and api-schema.yaml formats from your Tech Spec, ensuring each task has clear input/output specifications and dependencies for coordinated execution.

Can I use this to prepare PRDs for multi-agent execution?

Yes. The Skill transforms PRDs and Tech Specs into CPO-ready artifacts including manifest.json, task graphs, dependency declarations, and generated prompts. This prepares your specification for parallel multi-agent processing with defined waves and task ordering.

What files and formats does the decomposition workflow produce?

The workflow generates manifest.json (CPO format), context.md, architecture.md (with Mermaid diagrams), contract files (types.py, api-schema.yaml), a task graph, and task-specific prompts. All artifacts organize your PRD into an executable parallel structure.

Do I need existing Tech Specs to decompose a PRD?

A Tech Spec accelerates decomposition by providing architectural detail, but the Skill analyzes PRDs directly. Whether you start with a PRD alone or pair it with a Tech Spec, the workflow produces the same contract and task artifacts.

What's the difference between task waves and the task graph?

Waves group tasks for sequential execution phases, while the task graph shows all dependencies and parallel execution paths. The Skill generates both to clarify which tasks run in parallel within each wave and how work flows across stages.