Software Quality Assurance Analysts and Testers

/investigate — 버그 조사

버그 조사 스킬. 가설 3개를 먼저 세우고, 각각을 검증한 뒤, 사용자 승인 전까지 프로덕션 코드를 수정하지 않는 읽기 전용(freeze) 조사 모드. 이 스킬은 investigate, 버그, bug, debug, 조사, 디버깅, error, crash, 장애, incident, "왜 이러지", "원인 파악", "에러 나는데", "뭐가 문제야", "왜 실패하지", "이거 왜 안 돼", root cause, 근본 원인, "디버그 해줘" 등 버그/에러/장애 조사가 필요한 모든 상황에서 반드시 사용해야 한다. 테스트 실패를 보고 바로 코드를 수정하려 하기보다 먼저 이 스킬로 원인을 조사하라. mstack SDLC 파이프라인에서 /qa 실패 시 호출되며, 수정 후 /qa로 돌아간다.

코드 리뷰

Code and diff review skill for Codex. Use when code, a diff, or concrete implementation artifacts already exist and the main task is to classify findings into AUTO-FIX and SURFACE, spot missing tests, and compare the change against the plan; triggers include review, PR review, code review, and merge-readiness checks. Prefer `mstack-plan` for pre-implementation design and `mstack-pipeline` only when the user wants review chained with implementation, QA, shipping, and retro.

Safety Rules

Cross-cutting safety guardrail skill for Codex. Use when work could be destructive, expose secrets, touch shared modules, rewrite history, investigate failures under freeze, or change protected branches; triggers include "조심해서", safety, protection, secrets, rollback risk, and main-branch work. Prefer the task-specific mstack skill for planning, review, QA, shipping, or investigation after the safety posture is set.

Investigate

Failure diagnosis skill for Codex. Use when the main task is to reproduce a bug, form hypotheses, gather evidence, and identify the root cause before editing production code; triggers include crash, failure, regression, root cause, and "왜 안 돼" requests. Prefer `mstack-qa` for verification-only work, `mstack-plan` for pre-implementation design, and `mstack-pipeline` only when the user wants the full fix-through-release workflow.

CI/CD

Release Prep

Release gate skill for Codex. Use when review and QA evidence already exist and the main task is to validate tests, coverage, commit hygiene, branch strategy, and safe push behavior before merge or deployment; triggers include ship, release, push, merge, deploy, and handoff-before-merge requests. Prefer `mstack-plan` for design-only work, `mstack-qa` for verification-only work, and `mstack-pipeline` for one-shot end-to-end workflows.

Finance & Investment

Decision Router

Gate/Risk/STOP 결과로 AUTO_MERGE/PR_ONLY/ZERO_STOP 결정. decision/AUTO_MERGE/PR_ONLY/ZERO_STOP 키워드면 사용.

CI/CD

Release Prep

Autosort Policy Check

AGENTS.md(DEV-PRESET Autosort SSOT) 정책 준수 여부를 체크리스트로 점검. no delete, no dev rename, ledger, quarantine, temp stability, LLM 제한을 확인할 때 사용.

Git Workflows

Release Check

Run the minimum release-readiness checklist for this repo. Use before pushing a branch, opening a PR, or handing off work.

Automation Tools

Roborev Review

Run roborev code review on commits or uncommitted changes. Use to get AI-powered code review feedback.

machow0

Automation Tools

Roborev Show

Show existing code review results for a commit or job. Use to check review status without re-running.

machow0

E2e Testing

Write end-to-end tests using Playwright. Test user behavior, not implementation details. Also useful for non-test Playwright scripting.

lecstor0

Automation Tools

Review To Md

コードレビューをする

macinjoke0

Test Environment

Environment profile management for the Web Testing Agent. Use this skill when creating, loading, or checking environment profile.

maciej-szwarc-igt0

Security

Pm Verification Protocols

QA verification gate and evidence requirements

MacPhobos0

Mpm Postmortem

Analyze session errors and suggest improvements

MacPhobos0

Browse

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this". (gstack)

huichen0

Education

Specfact Code Review

House rules for AI coding sessions derived from review findings

nold-ai0

Review Code

Review code changes for quality and correctness. Use when: user wants any review of code changes.

mackeper0

Tdd Workflow

Guide for Test-Driven Development workflow. Use this when implementing features, writing tests, or following the red-green-refactor cycle.

lebobo880

Personal Test Writer

Write tests for a described feature or endpoint using the project's pytest, DRF, helpers, naming, and mocking conventions.

Macwdo0

Test Fixture V2

创建或更新 SQL 测试数据 fixture（单库、跨库、验证码登录场景），用于 nova 项目 API/集成测试的可重复数据准备。用户提到 fixture、测试数据造数、SQL 造数、登录/验证码相关测试、需要幂等可重复执行的测试准备时触发。

LeapBound0

CI/CD

Tdd Workflow

TDD red-green-refactor workflow with 5-step process and quality gates.

lebobo880

Code Reviewer

Review code for quality, correctness, and adherence to WikiSmith conventions. Use when asked to review code, PRs, or check for style/architecture issues.

macseem0

Django Tdd

Django testing strategies with pytest-django, TDD methodology, factory_boy, mocking, coverage, and testing Django REST Framework APIs.

Macwdo0

Testeur QA — Pipeline Orchestré avec Agent Teams

Ce skill orchestre la phase Testeur QA en 5 phases : analyse des user stories existantes, transformation en tests YAML exécutables, audit pré-déploiement via reverse-prompt (6 angles d'attaque), exécution parallèle avec Agent Teams (1 browser-qa-agent par story), et rapport agrégé avec screenshots. Le lead coordonne les agents QA, ne teste jamais directement. Use when l'utilisateur veut tester l'application, lancer les tests QA, exécuter les user stories, ou auditer le code avant déploiement.

leadmind-ai0

Run pre-commit quality checks and verify tests, formatting, linting, and documentation before PR creation.

lealvona0

Run pre-commit quality checks and verify tests, formatting, linting, and documentation before PR creation.

lealvona0