qa

Documents resolved implementation questions and decisions in qa.md before planning begins.

4|Updated Jan 1, 2026
One-click install
npx skills add https://github.com/vertexcover-io/tarash --skill qa-vertexcover-io
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qa
Source: https://github.com/vertexcover-io/tarash/tree/main/.claude/skills/qa
Command: npx skills add https://github.com/vertexcover-io/tarash --skill qa-vertexcover-io

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Bridges the gap between spec-generation and planning by surfacing implementation questions and documenting decisions so the planner starts with zero ambiguity.

Core Features & Use Cases

  • Interactive, step-by-step Q&A with the user to surface and resolve implementation questions before planning.
  • Deep codebase exploration using parallel agents to identify touched files, data flow, and design patterns.
  • Automatic generation of qa.md in the spec directory after all questions are resolved.
  • Seamless integration with the orchestrate pipeline to ensure a clean handoff from spec to plan.
  • Clear gating: announce at start and enforce a hard gate that requires interactive resolution before writing qa.md.

Quick Start

Point me to the spec.md (or the spec directory) and I will begin an interactive Q&A to resolve all design questions.

Frequently Asked Questions about qa

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

FAQPage Schema
How do I resolve ambiguous design requirements before starting implementation planning?

To resolve ambiguous design requirements before planning, use interactive low-level design QA to surface and document implementation decisions. It bridges spec-generation and planning by guiding step-by-step discussions that finalize architecture, interfaces, and acceptance criteria, ensuring you start planning with zero open questions.

What is the best way to document architecture decisions for a codebase with unclear requirements?

The best way to document architecture decisions for codebases with unclear requirements is through an interactive QA workflow. Deep codebase exploration identifies touched files and data flow, while the interactive discussion finalizes designs and automatically generates a qa.md file recording all resolved decisions.

How do I start an interactive Q&A session to finalize low-level design and acceptance criteria?

To start an interactive Q&A session for low-level design, point the tool to your spec.md or spec directory. It enforces a hard gate requiring interactive resolution of implementation questions, exploring the codebase to identify design patterns before writing the final qa.md document.

Can I use codebase analysis to identify touched files and data flow before writing a specification?

Yes, you can use codebase analysis to identify touched files, data flow, and design patterns before finalizing a specification. Parallel agents explore the codebase during the interactive QA process to surface implementation questions and ensure the resulting architecture decisions are grounded in existing code.

Does interactive design QA integrate with orchestration pipelines for spec-to-plan transitions?

Yes, interactive design QA integrates seamlessly with orchestrate pipelines to ensure a clean handoff from spec to plan. By documenting all resolved decisions in qa.md, it satisfies the pipeline requirement of starting the planning phase with zero ambiguity or open questions.

When should I not use interactive questioning for low-level design documentation?

You should not use interactive questioning for low-level design documentation when your requirements are already unambiguous and fully specified. The tool enforces a hard interactive gate to resolve open questions, making it unnecessary overhead for codebases with clear, finalized architecture and interfaces.