qa

Verify release behavior by running the product and reporting reproducible findings.

5|Updated Jan 31, 2026
One-click install
npx skills add https://github.com/nuggocto/dotfiles --skill qa-nuggocto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qa
Source: https://github.com/nuggocto/dotfiles/tree/main/opencode/skills/qa
Command: npx skills add https://github.com/nuggocto/dotfiles --skill qa-nuggocto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Passing automated tests and code review do not prove a feature actually works for real users. This Skill closes that gap by running the product end to end, exercising edge cases and failure states, and producing an evidence-backed release verdict. ## Core Features & Use Cases - End-to-end verification: Exercise the feature as a real supported user, then probe boundaries, state, permissions, failure recovery, and lifecycle paths with time-boxed exploratory charters. - Structured verdicts and reporting: Report PASS, PASS WITH KNOWN ISSUES, FAIL, BLOCKED, or INCONCLUSIVE alongside a separate ship/hold recommendation, with reproduction steps and evidence for every finding. - Disciplined intermittency handling: Preserve first failures, forbid retry-until-green, and require a stated diagnostic protocol for flaky results. - Use Case: Before shipping a new checkout flow, use this Skill to run the flow on staging with synthetic data, test double-submit and payment-timeout behavior, inspect logs and database state, and deliver a release verdict with documented residual risk. ## Quick Start Use the qa skill to verify the new checkout flow on staging and report a release verdict with reproduction steps for any issues found.

Frequently Asked Questions about qa

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

FAQPage Schema
How do I verify a feature works before release?

Run the product end to end as a real user, then probe edge cases, failure states, and permissions against acceptance criteria derived from requirements. Record the build, environment, and evidence, and report a verdict of PASS, FAIL, BLOCKED, or INCONCLUSIVE with reproduction steps.

What is the difference between QA verification and automated testing?

Automated tests check coded expectations, while QA verification observes actual release behavior including side channels like logs, database state, and network responses. Passing tests are relevant evidence but insufficient proof that a feature is fit to ship.

How should I handle flaky or intermittent test results?

Preserve the first failure and never retry until green. Repeat only under a stated diagnostic protocol on the same build, record every outcome, and report INCONCLUSIVE while a release-relevant intermittent result remains unexplained.

Can QA testing be done in production?

Production testing requires explicit authorization for the exact actions and effects. Prefer an isolated local or staging environment, and never trigger real payments, emails, destructive writes, or third-party effects without explicit approval.

When should QA not claim security or performance coverage?

QA may run safe acceptance checks but must not claim a security, accessibility, or performance audit unless that specialist scope was exercised. Load dedicated security, web-perf, or benchmark skills for those audits.