
Qaskills
io.github.PramodDuttaPublished by GitHub account PramodDuttav0.1.2Updated Oct 24, 2026
Search, inspect, and install 400+ QA testing skills from qaskills.sh into any MCP client.
Overview
A local MCP server that lets an AI assistant search, read and install curated QA testing skills from qaskills.sh into its own agent configuration.
- What you can do
- Search and browse the QA skills directory, including categories and a leaderboard of skills and their quality scores. Read a skill's details and full SKILL.md content before installing it. Install a chosen QA skill into the detected AI agent's configuration directory. List the categories available in the directory. Skills cover areas such as Playwright, Cypress, Jest, API, security, performance and accessibility testing.
- When to use it
- Useful when you want your coding agent to gain expert QA testing patterns, such as Playwright end-to-end or API test conventions, without copying them by hand. Suited to teams that want a curated, QA-specific skill library rather than generic skill collections.
- Requirements
- Runs locally over stdio, typically launched with npx from the npm package @qaskills/mcp, so Node.js and network access to qaskills.sh are needed. The optional QASKILLS_API_URL environment variable overrides the API base URL. No authentication is declared.
Installation
In SourceWeft
- Open Qaskills in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
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.
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)
2. Or install globally for faster access
3. Search for skills
4. List installed skills
5. Remove a skill
6. Or use the MCP server (search and install from inside your agent)
Listed in the official MCP registry as io.github.PramodDutta/qaskills.
Your agent gets six tools: search_skills, get_skill, get_skill_content, install_skill, list_categories, get_leaderboard. See packages/mcp.
Feature Highlights
Supported Frameworks
Supported AI Agents (27+)
Available Skills
Skill Packs
Install multiple skills at once with curated packs:
How It Works
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
- Go to qaskills.sh/dashboard/publish
- Sign in with GitHub
- Fill in the 5-step form
- Publish!
Via the CLI
SKILL.md Format
Tech Stack
Project Structure
Contributing
We welcome contributions! Here's how to get involved:
- Add a new skill — Create a
SKILL.mdand publish via the dashboard or CLI - Improve existing skills — Submit PRs to enhance testing patterns and coverage
- Report bugs — Open an issue on GitHub
- Feature requests — Open an issue or start a discussion
Local Development
Built by The Testing Academy
189K+ YouTube Subscribers | QA & Test Automation Education
MIT License | Made with testing in mind
Source: README.md at commit aada2bc
Tools
0Version history
1- v0.1.2LatestSep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts