spec-driven-development

Run a 5-phase gated pipeline from specifications to verified code.

40|6|Updated Jul 11, 2026
One-click install
npx skills add https://github.com/magnus919/agent-skills --skill spec-driven-development-magnus919
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-driven-development
Source: https://github.com/magnus919/agent-skills/tree/main/spec-driven-development
Command: npx skills add https://github.com/magnus919/agent-skills --skill spec-driven-development-magnus919

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill solves the problem of inconsistent, low-quality AI code generation by enforcing a rigorous, gated pipeline where structured specifications act as the single source of truth.

Core Features & Use Cases

  • Gated Pipeline: Implements a 5-phase pipeline (Specify, Decompose, Implement, Verify, Deliver) with 4 quality gates to catch defects early.
  • Standardized Templates: Provides ready-to-use templates for specifications, task plans, reviews, and verification reports.
  • Validation Scripts: Includes automated bash scripts to validate spec quality and task coverage against acceptance criteria.
  • Use Case: Use this when building a complex feature where you need to ensure that AI agents produce code that is fully compliant with non-functional requirements, data contracts, and edge-case handling.

Quick Start

Load the spec-driven-development skill and run the spec-quality-check script on your draft SPEC.md file to validate that all required sections are present.

Frequently Asked Questions about spec-driven-development

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

FAQPage Schema
How do I enforce quality gates in AI code generation pipelines?

Quality gates in AI code generation pipelines are enforced through a 5-phase methodology using 4 verification gates that validate task coverage and spec compliance against acceptance criteria via automated bash scripts.

What is spec-driven development for AI agents?

Spec-driven development for AI agents is a software factory methodology where structured specifications act as the single source of truth and executable inputs for AI code generation, ensuring compliance with non-functional requirements and data contracts.

How do I validate a specification before AI code generation?

Validate a specification by running an automated bash validation script against your draft SPEC.md file to verify all required sections are present and task coverage meets acceptance criteria before implementation.

Do I need bash to run spec-driven development workflows?

Yes, bash is required to run the automated validation scripts that verify spec quality and task coverage against acceptance criteria throughout the gated pipeline phases.

What's the best way to decompose complex features for AI code generation?

The best way to decompose complex features for AI code generation is using a structured pipeline that breaks specifications into standardized task plans, ensuring AI agents handle edge cases and non-functional requirements correctly.