convez.ai
MCP ServerAutomaçãoIntermediário
aks-builds

Quality Skills para IA — 107 Skills MCP para QA Engineering

aks-builds/quality-skills

107 skills de QA production-grade com contexto de stack para agentes IA

Ver no GitHub
BomPluginScore · de 0 a 100

MCP server com 107 skills production-grade para engenharia de QA distribuídas entre automação frontend, testes de API, performance, mobile, segurança e CI/CD. Funciona nativamente com Claude Code, Claude Desktop, Cursor, Zed, Windsurf e qualquer agente MCP, fornecendo conhecimento especializado de ferramentas de teste para gerar testes que realmente funcionam em produção.

Instalação

$npx convez-ai aks-builds/quality-skills

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
5
Forks
1
Acompanhando
0
Saúde
85%
Último commit
há 4 semanas
Versão
—
Issues abertas
8
Contribuidores
1

Casos de uso

  • Automação de testes end-to-end com Playwright e Selenium
  • Testes de API, performance e segurança via CI/CD
  • Scaffolding de testes BDD, mobile e acessibilidade com IA
  • Orquestração de CI e chaos engineering

Veredito

MCP server robusto e bem estruturado para agregar conhecimento de QA a agentes IA, eliminando alucinações sobre CLI flags e padrões de teste. Especialmente valioso para times que querem automação inteligente alinhada à stack técnica real, embora exija configuração inicial clara de qa-context.

README

A tradução para o português deste README ainda está na fila.

🔬 quality-skills

Give your AI agent the QA instincts of a senior SDET.

106 production-grade skills covering the complete quality engineering ecosystem — frontend automation, API testing, performance, mobile, BDD, accessibility, security, chaos engineering, contract testing, CI orchestration, and quality strategy. Works with Claude Code, Claude Desktop, Cursor, Zed, Windsurf, and any agent that speaks MCP or the Agent Skills spec.

Validate Skill Evals Skills: 106 MCP Agent Skills spec License: MIT

☝️ Claude Code invoking the playwright_scaffold tool — one of 260+ typed MCP tools in this collection.


Why quality-skills

AI coding agents are great at writing code — but without domain knowledge, they fabricate CLI flags, invent config keys, and generate test patterns that look right but break in CI. quality-skills gives your agent accurate, up-to-date knowledge of 40+ test automation tools and frameworks, so it helps you write tests that actually pass.

Every skill reads your project's qa-context first — your stack, languages, CI provider, browsers in scope — then applies the right tool idioms, not generic advice. The result: an agent that knows the difference between page.locator and page.find, understands k6 thresholds, and won't suggest Selenium when you're already on Playwright.

It works as a Claude Code plugin (one command to install), an MCP server (any MCP-compatible agent — Claude Desktop, Cursor, Zed, Windsurf, custom agents), or a plain file copy into .agents/skills/.

How skills work together

Skills reference each other and build on shared context. qa-context is the foundation — every other skill reads it first to understand your stack before doing anything.

                        ┌───────────────────────────────┐
                        │           qa-context          │
                        │  (read by all skills first)   │
                        └───────────────┬───────────────┘
                                        │
   ┌──────────┬───────────┬─────────────┼─────────────┬───────────┬──────────┐
   ▼          ▼           ▼             ▼             ▼           ▼          ▼
Strategy  Frontend     API          Performance   Mobile     A11y &     CI &
          Automation   Testing      & Load        Testing    Visual     Infra

See each skill's Related Skills section for the full dependency map.

Install

Option 1 — Claude Code Plugin (recommended)

/plugin marketplace add aks-builds/quality-skills
/plugin install quality-skills

Option 2 — MCP Server (Claude Desktop, Cursor, Zed, Windsurf, custom agents)

Add to your mcp_servers config (e.g. .mcp.json or claude_desktop_config.json):

{
  "mcpServers": {
    "quality-skills": {
      "command": "npx",
      "args": ["github:aks-builds/quality-skills/mcp"]
    }
  }
}

Or if you have it cloned locally:

{
  "mcpServers": {
    "quality-skills": {
      "command": "node",
      "args": ["/path/to/quality-skills/mcp/index.js"]
    }
  }
}

Option 3 — Clone and Copy

git clone https://github.com/aks-builds/quality-skills.git
cp -r quality-skills/skills/* .agents/skills/

Option 4 — Git Submodule

git submodule add https://github.com/aks-builds/quality-skills.git .agents/quality-skills

Usage

Once installed, ask your agent naturally:

"Set up Playwright for our Next.js app with parallel execution and tracing"
→ playwright_scaffold + ci-test-orchestration

"Why are these Cypress tests flaky?"
→ cypress + flaky-test-management

"Build a k6 load test for our checkout API targeting 200 RPS"
→ k6_scaffold + qa-context

"Add Pact contract testing between our web app and orders API"
→ pact-contract-testing + rest-assured

"Write a Cucumber feature for the password reset flow"
→ cucumber-gherkin + bdd-anti-patterns

Or invoke skills directly:

/playwright
/k6
/accessibility-testing
/hipaa-compliance

MCP tool calls (any MCP host):

playwright_scaffold({ target: "https://app.example.com/login", language: "ts" })
playwright_debug({ error: "TimeoutError: locator.click timed out after 30000ms" })
k6_scaffold({ endpoint: "/api/checkout", vus: 50, duration: "2m" })

Available Skills

Skill Categories

Foundation

  • qa-context — Stack, languages, CI provider, environments, quality bar — every other skill reads this first

Strategy & Fundamentals

  • test-strategy · test-design-techniques · test-data-management · flaky-test-management · test-environment-management

Frontend / Browser Automation

  • selenium · cypress · playwright · webdriverio · puppeteer · testcafe
  • New: testing-library · msw · storybook-testing · playwright-component · web-vitals · visual-ai-testing · browser-compatibility-testing · pwa-testing · electron-testing

Mobile Automation

  • appium · espresso · xcuitest · detox · maestro
  • New: flutter-testing · react-native-testing · mobile-performance-testing · mobile-accessibility-testing · desktop-testing

API / Backend Automation

  • postman-newman · rest-assured · supertest · pytest-api · karate · pact-contract-testing · wiremock · graphql-testing · grpc-testing
  • New: openapi-testing · websocket-testing · event-driven-testing · async-contract-testing · database-testing · api-fuzzing · schema-validation-testing · soap-xml-testing

Performance & Load

  • k6 · jmeter · gatling · locust · artillery

BDD & Specification

  • cucumber-gherkin · specflow-reqnroll · behave · bdd-anti-patterns

Unit Testing

  • jest-vitest · pytest · junit-testng · xunit-nunit · go-test · rspec
  • New: property-based-testing · snapshot-testing · test-doubles · parameterized-testing · approval-testing

Accessibility & Visual

  • accessibility-testing · visual-regression
  • New: accessibility-advanced · visual-ai-testing

Security & Resilience

  • security-testing · chaos-engineering · mutation-testing
  • New: sast-tooling · dast-tooling · dependency-scanning · secret-scanning · container-security-testing · compliance-testing

CI & Infrastructure

  • ci-test-orchestration · selenium-grid · cloud-test-grids · testcontainers · test-reporting · code-coverage
  • New: infrastructure-testing · serverless-testing · kubernetes-testing · test-selection · pipeline-quality-gates · performance-regression-testing

Specialized / Emerging

  • ai-augmented-testing · llm-eval-testing · feature-flag-testing · production-testing
  • New: ai-agent-testing · rag-testing · prompt-regression-testing · synthetic-data-ai · observability-testing · data-quality-testing · etl-testing · i18n-l10n-testing · test-metrics-dashboards · shift-right-testing · game-testing

Repository layout

quality-skills/
├── .claude-plugin/
│   ├── plugin.json            Claude Code plugin manifest
│   └── marketplace.json       Claude plugin marketplace entry
├── mcp/
│   ├── index.js               MCP server entrypoint
│   ├── loader.js              SKILL.md parser + skill-graph resolver
│   └── manifest.js            Generates tool manifest from all skills
├── skills/
│   └── skill-name/
│       ├── SKILL.md           Skill instructions (v2 frontmatter)
│       ├── evals/
│       │   └── evals.json     5–6 eval scenarios (CI-benchmarked)
│       └── references/        Optional deep-dive docs loaded on demand
├── AGENTS.md                  Guidelines for AI agents working in this repo
├── CLAUDE.md                  Claude Code reminders
├── CONTRIBUTING.md
├── LICENSE
├── README.md
├── SECURITY.md
└── VERSIONS.md

Contributing

PRs and issues welcome. See CONTRIBUTING.md — note the elevated bar for accuracy in tool-specific content (CLI flags, API signatures, and config keys must be verified against current docs before merging).

License

MIT — use these however you want, but verify before relying on them in production CI pipelines.

Ver o original no GitHub

Bom para

Relacionados