Claude Skills
CollectionsCompareWorkflowsNominate
Sign inSign up
© 2026 Curated Agent Skills·Learn more about Agent Skills
Back to repository

spec-authoring

verified

Expertise in writing clear, testable specifications. Activates when user discusses requirements, features, user stories, or acceptance criteria. Trigger keywords: specification, requirements, user story, acceptance criteria, feature spec, functional requirement, non-functional requirement, spec.md

View on GitHub

Marketplace

synapse-marketplace

datamaker-kr/synapse-claude-marketplace

Plugin

speckit-helper

development

Repository

datamaker-kr/synapse-claude-marketplace
1stars

plugins/speckit-helper/skills/spec-authoring/SKILL.md

Last Verified

February 5, 2026

Install Skill

Select agents to install to:

Scope:
npx add-skill https://github.com/datamaker-kr/synapse-claude-marketplace/blob/main/plugins/speckit-helper/skills/spec-authoring/SKILL.md -a claude-code --skill spec-authoring

Installation paths:

Claude
.claude/skills/spec-authoring/
Powered by add-skill CLI

Instructions

# Spec Authoring Skill

## Purpose

This skill provides expertise in writing clear, testable, and structured specifications. It activates automatically when the user discusses requirements gathering, feature definitions, user stories, or acceptance criteria. The goal is to produce specification documents that are unambiguous, complete, and directly translatable into implementation tasks.

## When It Activates

The skill is triggered when the conversation involves:

- Drafting or refining a new specification document (`spec.md`)
- Defining functional requirements (FR) or non-functional requirements (NFR)
- Writing user stories or acceptance criteria
- Discussing feature scope, constraints, or success metrics
- Reviewing specification quality or completeness

## Capabilities

### 1. Structured Specification Generation

Generate specifications following a consistent template structure:

- **Overview**: Project context, goals, and scope boundaries
- **Functional Requirements**: Numbered as `FR-001`, `FR-002`, etc.
- **Non-Functional Requirements**: Numbered as `NFR-001`, `NFR-002`, etc.
- **Data Model**: Entity definitions, relationships, and constraints
- **API Contracts**: Endpoint definitions with request/response schemas
- **Success Metrics**: Measurable criteria for feature validation

### 2. Requirement Numbering

All requirements follow a strict numbering convention:

- Functional requirements: `FR-001`, `FR-002`, `FR-003`, ...
- Non-functional requirements: `NFR-001`, `NFR-002`, `NFR-003`, ...
- Each requirement is atomic, testable, and traceable to user stories

### 3. User Story Decomposition

Break down features into user stories using the standard format:

```
As a [role],
I want to [action],
So that [benefit].
```

Each user story is labeled (e.g., `US1`, `US2`) and linked to one or more functional requirements.

### 4. Acceptance Criteria (Given/When/Then)

Every user story includes acceptance criteria in Gherkin-style syntax:

```
Given [precondition],
Wh

Validation Details

Front Matter
Required Fields
Valid Name Format
Valid Description
Has Sections
Allowed Tools
Instruction Length:
4175 chars