ultraqa

Run verification commands, diagnose failures, and apply fixes until the goal condition is met.

3|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ultraqa-marcmunoz-uno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ultraqa
Source: https://github.com/marcmunoz-uno/jailbreak/tree/main/skills/ultraqa
Command: npx skills add https://github.com/marcmunoz-uno/jailbreak --skill ultraqa-marcmunoz-uno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

UltraQA eliminates stalled or unreliable releases by repeatedly running the project’s QA checks, diagnosing failures, applying targeted fixes, and re-verifying until the stated quality goal is achieved.

Core Features & Use Cases

  • Goal-driven verification: Runs the appropriate check set for tests, build, lint, typecheck, or a custom success pattern based on your command arguments.
  • Autonomous diagnosis and repair loop: Sends failing output to an architect for root-cause analysis, then instructs an executor to apply precise file-level fixes and repeat.
  • Failure pattern control: Detects repeated failures (up to three times) and caps cycles (max 5) to avoid infinite loops while providing actionable stop reasons.
  • Stateful observability: Tracks cycle count, failure history, and session metadata in a local OMC state file for consistent progress reporting and cleanup.

Quick Start

Run the skill with the desired goal and checks, for example: /oh-my-claudecode:ultraqa --tests.

Frequently Asked Questions about ultraqa

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

FAQPage Schema
How do I automate running tests and fixing failures in a continuous integration workflow?

Automating QA in a continuous integration workflow involves running verification commands, diagnosing failures, and applying targeted fixes iteratively. This Skill cycles through checks for tests, build, lint, or typecheck until the goal condition is met.

How does automated root-cause analysis work for failing build and typecheck checks?

Automated root-cause analysis for failing checks sends the failing output to an architect component for diagnosis, then instructs an executor to apply precise file-level patches and repeat the verification cycle.

How do I prevent infinite loops when auto-fixing lint and test failures?

To prevent infinite loops during auto-fixing, the process detects repeated failure patterns up to three times and caps the total execution cycles at a maximum of five retries, providing actionable stop reasons.

Can I use this for custom output-pattern validation across software repositories?

Yes, you can use this for custom output-pattern validation across software repositories. It applies to CI-like workflows by running custom verification commands and validating success patterns based on command arguments.

What do I need to run iterative quality assurance checks for my project?

To run iterative quality assurance, you need a software repository with defined verification commands for testing, building, linting, or typechecking. The Skill tracks cycle count and failure history in a local OMC state file for cleanup.