qa-sweep

Coordinate parallel sub-agents with adversarial review and 95% confidence gates.

437|45|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/zaxbysauce/opencode-swarm --skill qa-sweep-zaxbysauce
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qa-sweep
Source: https://github.com/zaxbysauce/opencode-swarm/tree/main/.claude/skills/qa-sweep
Command: npx skills add https://github.com/zaxbysauce/opencode-swarm --skill qa-sweep-zaxbysauce

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured protocol to coordinate parallel sub-agents with mandatory adversarial review and a 95% confidence gate to ensure code quality and governance before completion.

Core Features & Use Cases

  • Parallel sub-agent implementation to accelerate work while maintaining coverage of edge cases.
  • Independent adversarial review: spawn a fresh agent to validate the work end-to-end.
  • Completeness verification: map every planned item to actual changes and verify all gaps are addressed.
  • Use Cases: complex feature development, debugging sessions, and maintenance tasks that require rigorous checks.

Quick Start

Apply the QA sweep protocol to a feature to enforce parallel development and independent review.

Frequently Asked Questions about qa-sweep

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

FAQPage Schema
How do I automate parallel QA with strict code review gates?

Automate parallel QA by coordinating sub-agents to implement tasks while enforcing independent adversarial review and a 95% confidence gate before closing. This ensures end-to-end verification and completeness for complex feature development.

What is an adversarial review process for software feature verification?

Adversarial review is an independent validation process where a fresh agent is spawned to evaluate implemented work end-to-end. It maps planned items to actual changes, verifies all gaps are addressed, and enforces strict gatekeeping before task completion.

How do I implement parallel sub-agent coordination for complex debugging sessions?

Implement parallel sub-agent coordination by applying a structured QA sweep protocol that accelerates work through specialized roles while maintaining edge case coverage. Mandatory gate-reviewed completion ensures rigorous checks during debugging and maintenance tasks.

Does parallel QA verification work for complex feature development requiring multiple specialized roles?

Parallel QA verification is designed specifically for complex feature development requiring multiple specialized roles. It applies structured protocols to coordinate sub-agents, enforcing completeness verification and independent validation before closing tasks.

What is the best way to enforce completeness verification during code review?

Enforce completeness verification by mapping every planned item to actual changes and verifying all gaps are addressed. Spawning a fresh agent for independent adversarial review ensures end-to-end validation and meets the 95% confidence gate before closing tasks.

When should I not use a 95% confidence gate for task completion?

A 95% confidence gate with parallel sub-agent coordination is intended for complex feature work, debugging sessions, and maintenance tasks requiring rigorous checks. It may introduce unnecessary overhead for simple, straightforward tasks that do not require strict gatekeeping.