kanban-refine

Convert vague backlog items into testable requirements through structured interviews.

173|42|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/cyanluna-git/cyanluna.skills --skill kanban-refine
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kanban-refine
Source: https://github.com/cyanluna-git/cyanluna.skills/tree/main/kanban-refine
Command: npx skills add https://github.com/cyanluna-git/cyanluna.skills --skill kanban-refine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Backlog items are often described in vague terms, making it hard for teams to align and implement. This skill guides structured interviews to convert rough descriptions into a concrete, testable requirements document with a clear goal, scope, acceptance criteria, and edge cases.

Core Features & Use Cases

  • Structured interview rounds that elicit missing details (WHAT, WHY, SCOPE, ACCEPTANCE, CONSTRAINTS, EDGE CASES, DEPENDENCIES)
  • A mandatory interview loop with question batching and round-based refinement
  • A standardized requirements template that includes Goal, Scope, Requirements, Acceptance Criteria, Constraints, and Edge Cases
  • Model routing to select the appropriate refiner and log actions for traceability
  • API-backed updates to tasks to save refined descriptions and metadata

Quick Start

Provide a rough backlog item to begin the interview-driven refinement and produce a ready-to-implement requirements document.

Frequently Asked Questions about kanban-refine

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

FAQPage Schema
How do I refine vague backlog items into actionable requirements?

You refine backlog items by running structured interview rounds that elicit missing details like goals, scope, acceptance criteria, and edge cases, then applying a standardized template to produce a testable requirements document.

What is a structured interview for agile backlog refinement?

A structured backlog interview batches questions across WHAT, WHY, SCOPE, ACCEPTANCE, CONSTRAINTS, EDGE CASES, and DEPENDENCIES categories to transform rough task descriptions into concrete, implementable requirements.

How do I add acceptance criteria and edge cases to a rough task description?

Adding acceptance criteria and edge cases requires a mandatory interview loop that prompts for specific conditions and boundary scenarios, then updates the backlog task with a standardized requirements template.

Does backlog refinement work with tasks already in progress?

Backlog refinement applies specifically to tasks in todo status, ensuring they have concrete goals, scope, and testable acceptance criteria before implementation begins.

How do I maintain traceability when updating refined requirements?

Traceability is maintained through model routing that logs refinement actions and API-backed updates that save the refined descriptions and metadata directly to the original backlog task.