specify

Convert plain-English feature requests into technical specifications with user stories and test plans.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/macollins27/guardrails-plugin --skill specify-macollins27
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: specify
Source: https://github.com/macollins27/guardrails-plugin/tree/main/skills/specify
Command: npx skills add https://github.com/macollins27/guardrails-plugin --skill specify-macollins27

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Plain-English feature requests often lack the precision and constraints needed for rapid, predictable development. This Skill converts those requests into a complete, unambiguous technical specification that a team can implement with minimal back-and-forth.

Core Features & Use Cases

  • Draft user stories, acceptance criteria, data models, edge cases, and testing strategies from a plain-English description.
  • Ensure alignment with project governance, constitution, and architecture documents so the spec is compliant and implementable.
  • Produce a living spec that can be updated as requirements evolve or new constraints arise.

Quick Start

Describe your feature in plain English and I will convert it into a detailed, unambiguous technical specification.

Frequently Asked Questions about specify

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

FAQPage Schema
How do I convert a plain-English feature request into a technical specification?

A feature specification includes user stories, acceptance criteria, data models, edge cases, and test plans. It transforms vague requests into rigorous, unambiguous technical directives ready for development teams.

How do I ensure my technical spec aligns with project architecture and governance documents?

To ensure your technical spec aligns with project architecture and governance documents, the specification process analyzes your project's constitution, architecture constraints, and governance requirements to produce compliant, directly testable outputs.

What is the best way to generate acceptance criteria and test plans from a feature description?

The best way to generate acceptance criteria and test plans from a feature description is to parse the plain-English input to identify scope and edge cases, producing a rigorous specification designed to be directly testable by development teams.

Can I use this to create a living spec that updates as requirements evolve?

Yes, you can use this to create a living spec that updates as requirements evolve. The output specification is designed to be modified as new constraints arise, ensuring your technical documentation remains continuously accurate.

Does generating a feature spec require predefined data models and architecture inputs?

Generating a feature spec does not strictly require predefined data models as inputs, but supplying project architecture and governance documents ensures the output specification is fully compliant and implementable with minimal back-and-forth.