Claude Plugins for Testing and QA: 8 Developer Tools for Quality Assurance

The Claude Plugins Community repository includes eight testing and QA plugins: 42crunch-api-security-testing, ab-testing, meticulous, accessibility-test, agentsqa, testdino (with seven sub-skills), and figma-test.

These plugins help developers audit APIs, run visual regression tests, ensure accessibility compliance, and manage automated testing workflows directly within Claude Code. This guide covers each plugin's purpose, source location, and practical usage.

Security and API Testing Plugins

42crunch-api-security-testing

The 42crunch-api-security-testing plugin audits OpenAPI specifications against security standards. It runs live conformance checks and identifies OWASP API vulnerabilities including BOLA (Broken Object Level Authorization) and BFLA (Broken Function Level Authorization).

As defined in .claude-plugin/marketplace.json (lines 54-58), this plugin integrates 42Crunch's security-scanning capabilities into Claude's conversational interface.

# Run a full API-security audit

/api-security-testing audit --openapi ./spec.yaml

ab-testing

The ab-testing plugin generates A/B testing infrastructure and analytics. Found in .claude-plugin/marketplace.json (lines 88-93), it streamlines experiment setup without requiring external tooling.

# Spin up a new A/B experiment

/ab-testing create --name "LandingPageV2" --variations 2

Visual and UI Testing Plugins

meticulous

Meticulous provides visual regression testing for web and mobile applications. Categorized explicitly as testing in .claude-plugin/marketplace.json (lines 101-105), it enables developers to:

• Review test run results
• Investigate session replays
• Debug visual diffs between versions

# Review the latest visual-regression run

/meticulous review --run-id 1234

figma-test

The figma-test plugin generates UI tests from Figma design files. Listed in .claude-plugin/marketplace.json (lines 8408-8410), it bridges design handoff with automated testing.

# Produce UI tests from a Figma file

/figma-test generate --file mydesign.fig

Accessibility and Exploratory Testing

accessibility-test

The accessibility-test plugin creates WCAG-compliant accessibility tests for any user interface. Defined in .claude-plugin/marketplace.json (lines 218-223), it automates accessibility validation without manual audit overhead.

# Generate WCAG tests for a URL

/accessibility-test generate --url https://myapp.com

agentsqa

agentsqa is an AI-powered QA testing agent that explores applications like a real user. According to .claude-plugin/marketplace.json (lines 769-774), it discovers:

• Visual bugs
• Logical errors
• Accessibility problems missed by unit tests

# Start an exploratory QA session on a live site

/agentsqa explore --url https://myapp.com

TestDino Playwright Management Suite

The testdino plugin is the most comprehensive testing tool in the repository. Declared in testdino/.claude-plugin/plugin.json (lines 2-9), it provides a complete Playwright test management system through seven specialized skills.

Core testdino Skills

Skill	Purpose	Source Path
testdino-sessions	Manage test sessions	testdino/skills/testdino-sessions/SKILL.md (lines 2-8)
testdino-runs	Inspect Playwright test runs	testdino/skills/testdino-runs/SKILL.md
testdino-releases	Handle release workflows	testdino/skills/testdino-releases/SKILL.md
testdino-manual-tests	Manage manual test cases	testdino/skills/testdino-manual-tests/SKILL.md
testdino-manual-runs	Execute manual test runs	testdino/skills/testdino-manual-runs/SKILL.md
testdino-health	Check system health status	testdino/skills/testdino-health/SKILL.md
testdino-audit	Run TestDino audits	testdino/skills/testdino-audit/SKILL.md

Each skill is defined in its own SKILL.md file under testdino/skills/, following the repository's skill-based architecture.

# List the last 5 Playwright runs

/testdino-runs list --limit 5

# Run a Playwright audit on a test file

/testdino-audit run --file tests/e2e/login.spec.ts

How Testing Plugins Are Organized

The repository structure separates testing functionality across two layers:

1. Marketplace entries in .claude-plugin/marketplace.json — single-file plugin definitions for simpler tools
2. Full plugin directories like testdino/ — multi-skill plugins with dedicated skill files

Non-testing plugins like quickdesign (quickdesign/skills/quickdesign/SKILL.md) and eli5 (eli5/.claude-plugin/plugin.json) follow the same patterns but focus on design and explanation rather than quality assurance.

Installing and Using Testing Plugins

All plugins require installation in Claude Code before use. The repository does not specify version requirements in the analyzed files, so compatibility depends on your Claude Code version.

Commands follow a consistent slash-prefix pattern: /plugin-name action --flags. Skills within multi-skill plugins use hyphenated names: /testdino-sessions, /testdino-runs, etc.

Summary

• Eight testing and QA plugins are available: 42crunch-api-security-testing, ab-testing, meticulous, accessibility-test, agentsqa, figma-test, and the seven-skill testdino suite
• Security testing: 42crunch audits OpenAPI specs for OWASP vulnerabilities
• Visual testing: meticulous handles regression testing; figma-test generates UI tests from designs
• Accessibility: accessibility-test ensures WCAG compliance; agentsqa finds accessibility bugs through AI exploration
• Playwright management: testdino provides complete test lifecycle management through individual skills defined in testdino/skills/*/SKILL.md
• Source files: plugin definitions live in .claude-plugin/marketplace.json or dedicated plugin.json files with skills in SKILL.md files

Frequently Asked Questions

What is the most comprehensive testing plugin in the Claude Plugins Community repository?

testdino is the most comprehensive, offering seven specialized skills for Playwright test management: sessions, runs, releases, manual tests, manual runs, health checks, and audits. Each skill is defined in its own SKILL.md file under testdino/skills/.

How does agentsqa differ from traditional automated testing?

agentsqa uses AI to explore applications like a real user, finding visual bugs, logical errors, and accessibility problems that traditional unit tests miss. Unlike scripted tests, it adapts its exploration based on actual UI behavior.

Where are plugin definitions stored in the repository?

Simple plugins are defined in .claude-plugin/marketplace.json. Complex multi-skill plugins like testdino have dedicated directories with a plugin.json manifest and individual SKILL.md files for each capability.

Can I use these plugins without external testing infrastructure?

Some plugins require external services—42crunch-api-security-testing needs 42Crunch integration, meticulous requires Meticulous setup, and testdino expects a TestDino environment. Plugins like ab-testing, accessibility-test, and agentsqa may operate with fewer external dependencies depending on implementation.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}