spec

Generate a PRD from a structured interview with project-specific sections.

71|17|Updated Apr 24, 2026
One-click install
npx skills add https://github.com/AgentiveStack/skills --skill spec-agentivestack
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec
Source: https://github.com/AgentiveStack/skills/tree/main/spec
Command: npx skills add https://github.com/AgentiveStack/skills --skill spec-agentivestack

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The spec unit helps teams quickly gather and align on a feature's scope by conducting a focused interview that grounds the PRD in the project’s domain model, bounded contexts, and ubiquitous language.

Core Features & Use Cases

  • Structured interview workflow that surfaces problem statements, success criteria, and constraints.
  • Generates a formal spec document with sections for problem, solution, user stories, decisions, tests, and scope.
  • Syncs with domain/docs (e.g., CLAUDE.md, CONTEXT maps) to ensure alignment across contexts.

Quick Start

Initiate a feature spec session with the AI to generate a complete PRD.

Frequently Asked Questions about spec

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

FAQPage Schema
How do I generate a PRD grounded in a domain model?

To generate a PRD grounded in a domain model, use an AI-guided interview process that captures problem statements and constraints, producing a structured feature spec with sections for user stories, testing strategy, and scope.

What is the best way to write a feature spec for cross-functional teams?

The best way to write a feature spec for cross-functional teams is to use a guided interview workflow that surfaces problem statements, success criteria, and constraints, syncing with domain docs to ensure alignment across bounded contexts.

How do I create a structured PRD with user stories and testing strategy?

Create a structured PRD by initiating a feature spec session that conducts an interview and outputs a formal document with dedicated sections for problem, solution, user stories, decisions, tests, and scope.

When do I need to produce a feature spec using ubiquitous language?

You need to produce a feature spec using ubiquitous language when starting a new feature, enhancement, or significant system change across bounded contexts, ensuring the documentation aligns with the project's domain model.

Does the generated feature spec save to a specific directory structure?

Yes, the generated feature spec saves to a specific directory structure at docs/contexts/<primary-context>/specs/<slug>.md, organizing documentation by bounded context for easy retrieval and domain alignment.