spec

Manage versioned living specifications with YAML frontmatter and EARS-based requirements.

7|2|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/tomasmihalyi/living-spec-skill --skill spec-tomasmihalyi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec
Source: https://github.com/tomasmihalyi/living-spec-skill/tree/main/spec
Command: npx skills add https://github.com/tomasmihalyi/living-spec-skill --skill spec-tomasmihalyi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Consolidate and continuously maintain living specifications for Claude Code, enabling AI-assisted organization, drift detection, and unified governance across features and phases.

Core Features & Use Cases

  • Centralized living specs that stay in sync with code and design artifacts.
  • AI-guided workflow for creating, updating, and validating specifications using EARS-based requirements.
  • Drift detection and tiered approval to manage changes through autonomous, notify, and blocking gates.

Quick Start

Load the current CLAUDE.md project context and initialize the living spec workflow.

Frequently Asked Questions about spec

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

FAQPage Schema
How do I maintain living specifications that stay in sync with code changes in Claude Code?

You maintain living specifications by loading the CLAUDE.md project context and initializing a workflow that centralizes specs into a .specs directory, using AI-guided drift detection to keep documents aligned with code changes.

What is EARS-based requirement syntax and how does it help manage software specifications?

EARS-based requirements provide a structured syntax for writing unambiguous specifications, enabling AI-guided workflows to create, update, and validate feature requirements with consistent governance across project phases.

How do I detect specification drift when code and design artifacts change?

Drift detection is handled by an AI-assisted workflow that compares centralized living specs against current code and design artifacts, flagging misalignments and routing updates through tiered approval gates.

Can I use living specs governance for a large product team with multiple feature phases?

Yes, living specs governance supports product teams and software engineers by providing unified specification management across multiple features and phases, routing changes through autonomous, notify, and blocking approval gates.

What is the best way to organize versioned specifications in an AI development lifecycle?

The best approach is centralizing versioned specifications into a .specs directory using YAML frontmatter and markdown structures, enabling AI-assisted organization and continuous drift detection throughout the AI development lifecycle.

How do I start generating AI-assisted specifications for a new feature?

To start generating specifications, load your current CLAUDE.md project context and initialize the living spec workflow, which routes AI-guided creation and updates into the centralized .specs directory.