sdd-work-decomposition

Translate reviewed specifications into dependency-aware task plans with acceptance criteria.

86|9|Updated Jun 5, 2026
One-click install
npx skills add https://github.com/magnus919/hermes-profiles --skill sdd-work-decomposition
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-work-decomposition
Source: https://github.com/magnus919/hermes-profiles/tree/main/skills/sdd-work-decomposition
Command: npx skills add https://github.com/magnus919/hermes-profiles --skill sdd-work-decomposition

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Transforms reviewed specifications into an executable, dependency-aware task plan that traces back to the source spec and outputs structured planning artifacts.

Core Features & Use Cases

  • Generats a TASK-PLAN.md that documents phases, groups, and individual tasks with acceptance criteria linked to SPEC.md.
  • Creates CLAUDE.md directives and AGENTS.md for deployment handoffs, enabling deterministic implementation by agents.
  • Maintains traceability by mapping each task to SPEC.md sections and its acceptance criteria, with a dependency graph for scheduling.

Quick Start

Provide SPEC.md and a reviewed specification to generate a complete task plan and agent handoff artifacts.

Frequently Asked Questions about sdd-work-decomposition

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

FAQPage Schema
How do I translate a reviewed specification into an executable task plan?

Spec-to-task decomposition is the process of breaking down a reviewed SPEC.md into atomic tasks with clear acceptance criteria. It produces a dependency-aware TASK-PLAN.md and a dependency graph for scheduling, ensuring traceability back to the source specification.

How do I break down a software specification into atomic tasks with dependency analysis?

To break down a specification into atomic tasks, generate a TASK-PLAN.md that documents phases, groups, and individual tasks. This includes a dependency graph for scheduling and maps each task's acceptance criteria directly to sections within the SPEC.md for full traceability.

What's the best way to set up deterministic implementation handoffs for AI agents from a spec?

Setting up deterministic implementation handoffs for AI agents requires generating CLAUDE.md directives and an AGENTS.md file from your reviewed specification. These artifacts define guarded handoffs, enabling implementation agents to execute the generated task plan deterministically.

Do I need a reviewed SPEC.md to generate a task plan with dependency mapping?

Yes, a reviewed SPEC.md is required to generate a task plan. The decomposition process relies on the reviewed specification to map each atomic task's acceptance criteria back to specific sections, ensuring full traceability and accurate dependency analysis.

Can I maintain traceability from implementation tasks back to the original specification sections?

Yes, traceability is maintained by mapping each generated atomic task and its acceptance criteria directly to corresponding sections of the SPEC.md. This ensures that every phase and group in the TASK-PLAN.md traces accurately back to the source specification.