create-specification

Create machine-readable software specifications with structured frontmatter and defined sections.

1|1|Updated Apr 16, 2026
One-click install
npx skills add https://github.com/iCesofT/awesome-ai --skill create-specification-icesoft
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-specification
Source: https://github.com/iCesofT/awesome-ai/tree/main/skills/01-planning-create-specification
Command: npx skills add https://github.com/iCesofT/awesome-ai --skill create-specification-icesoft

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Create precise architecture-ready specifications that clearly define requirements, constraints, interfaces, and data contracts to guide AI-assisted development and ensure consistent delivery across teams.

Core Features & Use Cases

  • Structured frontmatter with version, domain, triggers, role, and scope to codify the specification's identity.
  • Template-driven content: Introduction, 1) Definitions, 2) Requirements, 3) Interfaces & Data Contracts, 4) Acceptance Criteria, plus non-functional considerations.
  • Output in the /spec/ directory following the naming convention spec-[a-z0-9-]+.md to enable automated discovery, versioning, and traceability.

Quick Start

Provide a concise machine-readable specification for the target solution and save it to the /spec/ directory following the naming convention spec-[a-z0-9-]+.md.

Frequently Asked Questions about create-specification

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

FAQPage Schema
How do I generate machine-readable software specification documents with structured frontmatter?

To generate machine-readable software specification documents with frontmatter, this Skill creates architecture-ready specs in the /spec/ directory using the spec-[a-z0-9-]+.md naming convention, enforcing a schema for version, domain, triggers, role, and scope. It ensures consistent delivery and automated discovery across teams.

What is the best way to define requirements, interfaces, and data contracts for AI-assisted development?

The best way to define requirements, interfaces, and data contracts for AI-assisted development is using template-driven specifications. This Skill produces structured sections for Definitions, Requirements, Interfaces & Data Contracts, and Acceptance Criteria to clearly establish constraints and guide consistent solution delivery.

How do I create precise architecture specifications that support versioning and traceability?

Creating precise architecture specifications that support versioning and traceability involves applying a structured frontmatter schema and mandated naming conventions. This Skill outputs machine-ready documents to the /spec/ directory, enabling automated discovery and tracking across development workflows.

Can I use this specification template for non-functional requirements and acceptance criteria?

Yes, you can use this specification template for non-functional requirements and acceptance criteria. The Skill's template-driven content explicitly includes dedicated sections for Acceptance Criteria and non-functional considerations alongside Definitions, Requirements, and Interfaces.

Does this specification approach enforce a specific file naming convention and directory structure?

Yes, this specification approach enforces a strict file naming convention and directory structure. It saves all generated markdown documents under the /spec/ directory using the designated spec-[a-z0-9-]+.md pattern to enable automated discovery, versioning, and traceability.

What sections are included in an architecture-ready specification for software solutions?

An architecture-ready specification for software solutions includes structured frontmatter followed by Introduction, Definitions, Requirements, Interfaces & Data Contracts, and Acceptance Criteria sections. It also incorporates guidelines for formatting, versioning, traceability, and non-functional considerations.