gh-mll-lab-claude-plugin-marketplace-plugins-spec/commands/req.md

Specs - Requirements from JIRA Story

This command captures user stories and acceptance criteria in structured EARS notation. It does not implement anything, but helps to define the requirements clearly.

Usage

To create a new requirement, just type:

/spec:req <jira_issue_id>

What This Command Does

1. Retrieves the story with the given ID <jira_issue_id> from JIRA.
2. Checks if a folder with a filepath-safe version of <jira_issue_id> already exists under .specs, if not it creates it.
3. Checks if a file named requirements.md exists in the newly created folder, if not it creates one.
4. Turns the description of the JIRA story into a fully fleshed out requirements.md file in the appropriate format.
5. Asks the user for additional details to fill out the requirements. This is a repetitive process until all details are captured.
6. Once finished, it asks the user to review the requirements.md file.
7. If the user approves, it determines the changes to the original requirements and posts them as comment to the JIRA issue in the language of the original issue description.

Appropriate Format

<!-- markdownlint-disable-file MD024 -->
<!-- markdownlint-disable-file MD013 -->
<!-- markdownlint-disable-file MD040 -->

# Requirements Document

## Introduction

Describe the feature in a detailed paragraph.

## Requirements

### Requirement 1

**User Story** Describe the user story for requirement 1.

#### Acceptance Criteria

1. Describe the first acceptance criterion in EARS format.
2. Describe the second acceptance criterion in EARS format.
3. Describe the third acceptance criterion in EARS format.

### Requirement 2

**User Story** Describe the user story for requirement 2.

#### Acceptance Criteria

1. Describe the first acceptance criterion in EARS format.
2. Describe the second acceptance criterion in EARS format.

EARS format

EARS (Easy Approach to Requirements Syntax) notation provides a structured format for writing clear, testable requirements. Each acceptance criterion should be expressed in the following format:

WHEN [condition/event]
THE SYSTEM SHALL [expected behavior]

IF [unwanted condition/event],
THEN THE SYSTEM SHALL [system response/behavior]

WHILE [system state],
THE SYSTEM SHALL [system response/behavior]

WHERE [feature],
THE SYSTEM SHALL [system response/behavior]

Note these patterns can be combined for more complex scenarios.