ultraqa

Coordinate tester, architect, and executor roles to fix issues across tests, builds, linting, and type checks.

Updated Feb 15, 2026
One-click install
npx skills add https://github.com/leejaedus/pepcode --skill ultraqa-leejaedus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ultraqa
Source: https://github.com/leejaedus/pepcode/tree/main/skills/ultraqa
Command: npx skills add https://github.com/leejaedus/pepcode --skill ultraqa-leejaedus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates autonomous QA cycling to continuously test, verify, and fix software until a quality goal is met.

Core Features & Use Cases

  • Cycle-driven QA orchestration: coordinates qa-tester, architect, and executor roles to diagnose and fix issues across tests, builds, linting, and type checks.
  • Flexible goals & observability: supports tests, build, lint, typecheck, and custom goals with clear cycle progress and state tracking.
  • Stateful, safe workflow: includes state management, cancellation, and cleanup for reproducible QA loops.

Quick Start

Begin the ULTRAQA cycle by setting a goal and letting it run through up to five cycles until the goal is met.

Frequently Asked Questions about ultraqa

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

FAQPage Schema
How do I automate QA cycles for continuous testing and fixing?

Automate QA cycles by coordinating multi-agent orchestration across test, verification, and fix phases. The system iterates through tests, builds, linting, and type checks until the defined quality goal is met.

Can I run autonomous software testing with custom validation goals?

Autonomous software testing supports custom validation goals alongside standard tests, build, lint, and typecheck targets. You set the objective, and the cycle autonomously works toward achieving that specific quality state.

How does multi-agent orchestration work for software diagnosis and fixing?

Multi-agent orchestration coordinates qa-tester, architect, and executor roles to diagnose and resolve issues. Each role handles specific tasks within the iterative cycle to systematically clear build, lint, and test failures.

What is the maximum number of cycles for autonomous QA validation?

Autonomous QA validation runs through up to five cycles to meet the specified goal. State tracking and clear progress reporting ensure you can monitor each iteration until the quality standard is achieved.

Does the autonomous QA workflow support state management and cancellation?

The autonomous QA workflow includes stateful state management, cancellation, and cleanup capabilities. This ensures reproducible QA loops and safe termination when you need to halt or reset the ongoing validation cycle.