mcpField
Q
qaskills

qaskills

Local
github.com/PramodDuttaDeveloper Tools244Updated 1w ago

About qaskills

QASkills.sh is a curated directory of QA testing skills for AI coding agents, each packaged as a SKILL.md file. Its MCP server lets users search, inspect and install these skills from inside an agent, aimed at developers and QA teams who want testing guidance in their agents.

Key features

  1. 01Search QA skills by keyword
  2. 02Inspect skill details and full content
  3. 03Install skills from inside the agent
  4. 04List skill categories and leaderboard
  5. 05Detect the AI agent and install to its config directory

Use cases

  1. 01Have an agent write Playwright E2E tests with Page Object Model
  2. 02Find API or performance testing skills for an agent
  3. 03Manage testing skills across several MCP clients

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"pramoddutta-qaskills": {
"command": "npx",
"args": [
"-y",
"@qaskills/mcp"
],
"env": {
"QASKILLS_API_URL": "<QASKILLS_API_URL>"
}
}
}
}
Run commandnpx -y @qaskills/mcp
Environment variables to set
QASKILLS_API_URLOverride the QASkills API base URL (defaults to https://qaskills.sh)

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@qaskills/mcp0.1.2

Tools

6

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
sortenumNotrending | newest | quality | popular
agentstringNo
limitnumberNo
querystringNo
languagestringNo
frameworkstringNo
testingTypestringNo

Details

From the project README

QASkills.sh

The Curated QA Skills Directory for AI Coding Agents

The curated QA skills directory for Claude Code, Cursor, Copilot, Windsurf, and 27+ AI coding agents.

npm version License: MIT Website YouTube

Website | Skills Directory | Leaderboard | Get Started


Why QASkills.sh?

Among 49,000+ skills on Vercel's skills.sh, barely any are dedicated to QA testing. QASkills.sh fills that gap — a curated, QA-specific skills directory that gives your AI coding agent expert-level testing knowledge with a single command.

npx @qaskills/cli add playwright-e2e
  ✓ Detected agent: Claude Code
  ✓ Installing playwright-e2e v1.0.0...
  ✓ Skill installed successfully!

Your AI agent now writes Playwright tests using Page Object Model, fixtures, and best practices — automatically.


Quick Start

1. Install a skill (no global install needed)
npx @qaskills/cli add playwright-e2e
2. Or install globally for faster access
npm i -g @qaskills/cli
qaskills add playwright-e2e
3. Search for skills
npx @qaskills/cli search "api testing"
4. List installed skills
npx @qaskills/cli list
5. Remove a skill
npx @qaskills/cli remove playwright-e2e
6. Or use the MCP server (search and install from inside your agent)

Listed in the official MCP registry as io.github.PramodDutta/qaskills.

# Claude Code
claude mcp add qaskills -- npx -y @qaskills/mcp
// Any MCP client (Cursor, Windsurf, etc.)
{
  "mcpServers": {
    "qaskills": { "command": "npx", "args": ["-y", "@qaskills/mcp"] }
  }
}

Your agent gets six tools: search_skills, get_skill, get_skill_content, install_skill, list_categories, get_leaderboard. See packages/mcp.


Feature Highlights

FeatureDescription
27+ Agents SupportedClaude Code, Cursor, Copilot, Windsurf, Codex, Aider, Cline, Devin, and more
Auto-DetectionCLI detects which AI agent you're using and installs to the right config directory
20+ Curated SkillsExpert-reviewed QA testing patterns covering E2E, API, unit, performance, security, and more
Skill PacksInstall bundles of related skills in one command
Publish Your OwnCreate and share skills via the dashboard or CLI
Open SourceMIT licensed, community-driven
Supported Frameworks
Playwright Cypress Selenium Jest Pytest k6 JMeter Appium Postman REST Assured Cucumber Axe-core Pact

Supported AI Agents (27+)

AgentStatusAgentStatus
Claude Code✅Cursor✅
GitHub Copilot✅Windsurf✅
Codex CLI✅Aider✅
Continue✅Cline✅
Zed AI✅Bolt.new✅
Lovable✅v0✅
Devin✅Sourcegraph Cody✅
Replit Agent✅Amazon Q Developer✅
JetBrains AI✅Gemini Code Assist✅
Roo Code✅Augment Code✅
Trae✅And more...✅

Available Skills

#SkillCategoryQuality
1Playwright E2E TestingE2E, Visual92/100
2Jest Unit TestingUnit91/100
3Cypress E2E TestingE2E90/100
4OWASP Security TestingSecurity89/100
5Playwright API TestingAPI88/100
6Pytest PatternsUnit, Integration88/100
7k6 Performance TestingPerformance87/100
8Visual Regression TestingVisual86/100
9Axe-core AccessibilityAccessibility86/100
10Selenium Java TestingE2E85/100
11REST Assured APIAPI85/100
12CI/CD Pipeline ConfigDevOps85/100
13Appium Mobile TestingMobile84/100
14Postman API TestingAPI84/100
15Contract Testing (Pact)Contract84/100
16Test Data GenerationUtility83/100
17JMeter Load TestingLoad82/100
18BDD/Cucumber PatternsBDD82/100
19Test Plan GenerationProcess80/100
20Bug Report WritingProcess78/100

Browse all skills →


Skill Packs

Install multiple skills at once with curated packs:

PackIncludesUse Case
Full Stack QAPlaywright + Cypress + k6 + OWASP + Jest + Pytest + CI/CDComplete QA coverage
Complete Playwright SuiteE2E + API + Visual + AccessibilityPlaywright-focused teams
API Testing ToolkitPlaywright API + REST Assured + Postman + Pact + k6Backend & API testing

How It Works

1. SEARCH   →  Browse skills on qaskills.sh or search via CLI
2. INSTALL  →  npx @qaskills/cli add <skill-name>
3. TEST     →  Your AI agent now has QA expertise built-in

The CLI auto-detects which AI agent you're using (Claude Code, Cursor, Copilot, etc.) and installs the skill's SKILL.md to the correct configuration directory. Your agent reads this file and follows the testing patterns automatically.


Publish Your Own Skill

Via the Website
  1. Go to qaskills.sh/dashboard/publish
  2. Sign in with GitHub
  3. Fill in the 5-step form
  4. Publish!
Via the CLI
# Scaffold a new SKILL.md
npx @qaskills/cli init my-skill

# Edit the generated SKILL.md with your testing patterns

# Publish to the registry
npx @qaskills/cli publish
SKILL.md Format
---
name: My Testing Skill
description: What this skill teaches AI agents
version: 1.0.0
license: MIT
testingTypes: [E2E, API]
frameworks: [Playwright]
languages: [TypeScript]
agents: [claude-code, cursor, copilot]
---

# My Testing Skill

Instructions for the AI agent go here...

Tech Stack

LayerTechnology
FrameworkNext.js 15 (App Router)
UIshadcn/ui + Tailwind CSS v4
DatabaseNeon Serverless Postgres + Drizzle ORM
SearchTypesense
AuthClerk
CacheUpstash Redis
HostingVercel
CLITypeScript + Commander.js
MonorepoTurborepo + pnpm

Project Structure

qaskills/
├── packages/
│   ├── web/              # Next.js 15 website & API
│   ├── cli/              # @qaskills/cli (npm package)
│   ├── shared/           # Types, constants, Zod schemas, parsers
│   ├── sdk/              # Programmatic TypeScript SDK
│   └── skill-validator/  # SKILL.md validation
├── seed-skills/          # 20+ curated QA skill definitions
└── landingpage/          # Marketing site

Contributing

We welcome contributions! Here's how to get involved:

  1. Add a new skill — Create a SKILL.md and publish via the dashboard or CLI
  2. Improve existing skills — Submit PRs to enhance testing patterns and coverage
  3. Report bugs — Open an issue on GitHub
  4. Feature requests — Open an issue or start a discussion
Local Development
# Clone the repo
git clone [email protected]:PramodDutta/qaskills.git
cd qaskills

# Install dependencies (requires pnpm 9.15+ and Node 20+)
pnpm install

# Build shared package first (other packages depend on it)
pnpm --filter @qaskills/shared build

# Start all dev servers
pnpm dev

# Run tests
pnpm test

# Lint & format
pnpm lint
pnpm format

Built by The Testing Academy

189K+ YouTube Subscribers | QA & Test Automation Education

YouTube | Website | Twitter


MIT License | Made with testing in mind

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.