What is the spec-methodology workflow?

It follows a four-step process: /spec:create to generate a document, /spec:validate to audit requirements, /spec:decompose to create tasks, and /spec:execute to implement via agents.

Can I use my own task management tools with this skill?

Yes, it integrates natively with STM (Session Task Manager) for persistent tracking and uses TodoWrite as a local session fallback.

How does this prevent overengineering?

The methodology includes a dedicated YAGNI (You Aren't Gonna Need It) check that identifies and cuts unnecessary features and complexity during the validation phase.

Why is 'Content Preservation' important in this skill?

It ensures that implementation details and code examples are copied verbatim from the spec into tasks, preventing loss of context during the development phase.

Specification-Driven Development

Name: Specification-Driven Development
Author: betamatt

bybetamatt

Productivity & Workflow

Implements a systematic specification-first workflow to eliminate overengineering and ensure code quality through first-principles analysis.

About

The spec-methodology skill brings Specification-Driven Development (SDD) to Claude Code, providing a structured four-phase framework: Create, Validate, Decompose, and Execute. By guiding users through first-principles problem analysis and rigorous YAGNI (You Aren't Gonna Need It) audits, it ensures that every line of code is backed by a validated requirement. This skill is particularly effective for complex features and bug fixes, as it enforces quality gates and ensures that implementation details are preserved verbatim when breaking down high-level specs into actionable, self-contained tasks for AI agents.

Key Features

0 GitHub stars
Integrated quality gates requiring high-scoring specs before implementation
Self-contained task decomposition with verbatim content preservation
Automated YAGNI auditing to aggressively reduce technical debt and scope creep
Four-phase SDD workflow: Create, Validate, Decompose, and Execute
First-principles problem analysis to identify core root causes

Use Cases

Managing multi-step AI agent execution through structured task decomposition
Architecting complex new software features from vague requirements
Refactoring legacy systems by first establishing a validated technical spec

About

Key Features

0 GitHub stars
Integrated quality gates requiring high-scoring specs before implementation
Self-contained task decomposition with verbatim content preservation
Automated YAGNI auditing to aggressively reduce technical debt and scope creep
Four-phase SDD workflow: Create, Validate, Decompose, and Execute
First-principles problem analysis to identify core root causes

Use Cases

Managing multi-step AI agent execution through structured task decomposition
Architecting complex new software features from vague requirements
Refactoring legacy systems by first establishing a validated technical spec