spec-driven-development

Convert vague ideas into structured specifications with spec.md, prompt_plan.md, and todo.md artifacts.

Updated Feb 5, 2026

Search Tags:#workflow #planning #documentation #specification #PRD #AI-assisted #frontmatter

Ready to run

npx skills add https://github.com/Norman-Norman-Norman/product --skill spec-driven-development-norman-norman-norman

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: spec-driven-development
Source: https://github.com/Norman-Norman-Norman/product/tree/main/.github/skills/spec-driven-development
Command: npx skills add https://github.com/Norman-Norman-Norman/product --skill spec-driven-development-norman-norman-norman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams convert vague ideas into structured, implementation-ready specifications for AI-assisted development.

Core Features & Use Cases

Draft comprehensive specs across the three-phase workflow (Idea Honing, Planning, Execution)
Generate supporting artifacts: spec.md, prompt_plan.md, todo.md
Provide clear guiding prompts and templates for AI agents to follow

Quick Start

To get started, provide your project idea and desired outcomes and follow the Skill's prompts to produce a complete specification, a detailed implementation plan, and a task checklist.

Frequently Asked Questions about spec-driven-development

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

FAQPage Schema

How do I turn a vague project idea into an executable specification for AI-assisted development?▼

To turn a vague idea into an executable specification, you provide your project concept and desired outcomes to a structured workflow that enforces frontmatter and generates implementation-ready PRDs, architecture notes, and testing plans.

What artifacts are generated when creating a PRD for AI coding agents?▼

Creating a PRD for AI coding agents generates three core artifacts: spec.md for the detailed specification, prompt_plan.md for guiding AI prompts, and todo.md for a structured task checklist.

Can I use this specification workflow for both greenfield and existing projects?▼

Yes, this specification workflow applies to both greenfield and existing projects, enforcing structured frontmatter and phased workflows to ensure clear requirements and measurable success criteria across any codebase.

What phases are involved in writing implementation-ready specifications?▼

Writing implementation-ready specifications involves a three-phase workflow: Idea Honing to refine concepts, Planning to establish architecture notes and success criteria, and Execution to generate detailed plans and task checklists for AI agents.

Do I need any specific dependencies to generate structured prompt plans for AI development?▼

No specific dependencies are required to generate structured prompt plans for AI development. The workflow operates independently to transform project ideas into clear instructions and testing plans.

Why should I use a phased workflow instead of writing a PRD directly for AI-assisted development?▼

Using a phased workflow for PRD creation ensures vague ideas are systematically refined into measurable success criteria and structured frontmatter, preventing misalignment between project requirements and AI agent execution outputs.