spec-driven

Create traceable feature specifications and execution plans for implementation workflows.

8|Updated Feb 9, 2026
One-click install
npx skills add https://github.com/adeonir/agent-skills --skill spec-driven-adeonir
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-driven
Source: https://github.com/adeonir/agent-skills/tree/main/skills/%28development%29/spec-driven
Command: npx skills add https://github.com/adeonir/agent-skills --skill spec-driven-adeonir

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Spec-driven development provides a disciplined, transparent workflow to convert ambiguous ideas into clear, traceable specs and executable plans, reducing miscommunication and rework.

Core Features & Use Cases

  • Establishes WHAT to build via spec.md (who, what, why) and acceptance criteria
  • Defines HOW to build via plan.md with architecture decisions, data models, and patterns
  • Breaks work into atomic tasks (tasks.md) and guides implementation, validation, and archiving
  • Supports greenfield and brownfield contexts, inline Q&A, and on-demand references

Quick Start

Run initialize to generate a new feature spec, plan, tasks, implement, validate, and archive.

Frequently Asked Questions about spec-driven

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

FAQPage Schema
What is spec-driven development and how does it structure feature planning?

Spec-driven development is a structured workflow that converts ambiguous ideas into traceable feature specifications and executable plans. It enforces structured frontmatter, open questions handling, data modeling, and quality gates to ensure clear decisions and repeatable outcomes.

How do I create traceable feature specifications and execution plans?

You create traceable feature specifications by running the initialize trigger to generate spec.md for requirements, plan.md for architecture decisions, and tasks.md for atomic task breakdowns. This workflow enforces structured frontmatter, open questions handling, data modeling, and quality gates.

Does spec-driven development work for brownfield codebases or only greenfield projects?

Spec-driven development supports both brownfield and greenfield contexts. For existing codebases, use the map codebase trigger to analyze current architecture before initializing new feature specs, execution plans, and atomic tasks.

How do I handle open questions and data modeling during feature planning?

Open questions and data modeling are handled through structured frontmatter enforcement and inline Q&A during the planning phase. The plan.md file captures architecture decisions, data models, and patterns to ensure clear decisions and repeatable outcomes.

What's the best way to guide implementation and validation after creating a feature plan?

The best way to guide implementation and validation is using the implement and validate triggers from the spec-driven workflow. These coordinate execution based on tasks.md atomic breakdowns and enforce quality gates before the archive trigger finalizes the feature.

Why does my feature development workflow lead to miscommunication and rework?

Miscommunication and rework occur when feature development lacks disciplined, transparent workflows. Spec-driven development solves this by establishing what to build via spec.md with acceptance criteria and how to build via plan.md with architecture decisions and data models.