sdd:plan

Orchestrate multi-agent planning workflows to refine draft tasks into implementation-ready plans.

Updated Mar 17, 2026

Search Tags:#workflow #multi-agent #plan #phases #task refinement #judges

Ready to run

npx skills add https://github.com/Avi977/ace-claude-toolkit --skill sdd-plan-avi977

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: sdd:plan
Source: https://github.com/Avi977/ace-claude-toolkit/tree/main/skills/spec-driven-development
Command: npx skills add https://github.com/Avi977/ace-claude-toolkit --skill sdd-plan-avi977

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Refine and orchestrate a structured, multi-agent workflow that turns draft tasks into implementation-ready plans and tasks.

Core Features & Use Cases

Orchestrates parallel analysis, architecture synthesis, decomposition, and verifications with judge checks.
Supports refine, continue, and human-in-the-loop modes, with configurable thresholds and stages.
Tracks artifacts and task transitions across draft → todo → in-progress → done.

Quick Start

Run the plan command on a draft task file to coordinate multi-agent refinement with integrated quality gates.

Frequently Asked Questions about sdd:plan

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

FAQPage Schema

How do I refine draft tasks into implementation-ready plans using a multi-agent workflow?▼

You can refine draft tasks by running a structured multi-agent planning workflow that applies parallel analysis and judge validations across configurable phases to produce implementation-ready tasks.

What is a multi-agent planning workflow with judge checks?▼

A multi-agent planning workflow coordinates parallel agents for analysis and architecture synthesis, applying judge validations as quality gates across configurable phases to rigorously refine draft tasks.

Can I configure phases and quality thresholds for task refinement and validation?▼

Yes, the workflow supports configurable phases with judge validations, customizable thresholds, and refine, continue, or human-in-the-loop modes to control the task refinement process.

Does the multi-agent planning workflow track task transitions from draft to done?▼

Yes, the workflow tracks artifacts and task transitions across draft, todo, in-progress, and done states to maintain visibility throughout the refinement process.

What's the best way to add human-in-the-loop reviews to an automated task planning pipeline?▼

You can add human-in-the-loop reviews by enabling continue or refine modes within the multi-agent planning workflow, which pauses execution for validation before moving tasks forward.

When should I not use a multi-agent planning workflow for task refinement?▼

Avoid multi-agent task refinement for simple tasks that do not require rigorous parallel analysis, architecture synthesis, or multi-phase judge validations, as the overhead outweighs the benefit.