Agent Plugins Marketplace
← All plugins

qa-engineering

v0.6.0

The single AI QA engineering plugin: one orchestrator for the whole loop — analyze a story or a module, author and approve test cases, build and generate API + UI automation, run the suite, and turn a failing run into a bug — human-gated at every tracker write. On first run it builds a frontend-first Knowledge Base of the product's features, flows and testable rules (discovered from the app's routes, deepened per module) that grounds story and module analysis. Publishing cases to a Test Plan/Suite and filing bugs support Azure DevOps today (other trackers planned); every other workflow is tracker-independent and works with no tracker at all. Also teaches and reviews a Playwright + TypeScript framework. Self-contained with its own native QA skills; ships no product data — the QA supplies tracker, API, auth and URLs locally through a git-ignored house-profile and .env, and the Knowledge Base is generated into the QA's own repo.

Claude Code4 Skills2 MCP serversstdio

By shironigun0 GitHub starsUpdated last week

Directory evidence

Runtimes
Claude Code
Parsed components
6 skill or MCP entries
Source updated
Oct 10, 2026
Manifest status
Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install qa-engineering for Claude Code

Installs for the current user
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install qa-engineering@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually
git clone https://github.com/shironigun/claude-marketplace

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/qa-engineering/.

Plugin files

plugins/qa-engineering/
├── .claude-plugin/plugin.json
├── skills/author-ui-cases/SKILL.md
├── skills/file-to-tracker/SKILL.md
├── skills/framework-standards/SKILL.md
├── skills/kit-builder/SKILL.md
└── .mcp.json

Included Skills4

author-ui-casesskills/author-ui-cases/SKILL.md

Native skill of the qa-engineering — review a user story, feature description, or acceptance criteria as a senior SQA analyst and write click-by-click UI test cases in a rigorous, behavior-driven house style, every step atomic, every step with an explicit expected result. Product-neutral and tracker-agnostic (ADO / Jira / any), grounded in the kit's framework standards. Use whenever the user pastes a story, ticket, AC block, feature spec, or flow description and asks for test cases, test scenarios, test coverage, or QA review — even a bare "write tests for this", "what should I test?", "break this story into test cases", or "give me QA coverage for this AC". Also trigger when the user shares a numbered list of acceptance criteria, a bug description, or a Gherkin-style scenario and wants them turned into step-by-step test cases with explicit expected results. Authors cases only; to file them as work items hand off to file-to-tracker.

file-to-trackerskills/file-to-tracker/SKILL.md

Router of the qa-engineering for the combined "write test cases for a ticket AND file them in the tracker" intent. It authors nothing and writes nothing itself — it sequences the two native engines — author the cases (author-ui-cases for UI, author-api-cases for API), which record them in qa-context with an approval seam, then file the approved set with the gated ado-publish (which picks the Test Plan/Suite, creates, adds-to-suite and links — nothing written until the QA confirms a dry-run). Product-neutral and tracker-agnostic (Azure DevOps is the implemented filing path; the same gated flow applies to Jira/any via that tracker's connector). Use whenever the user wants to write, draft, generate, or create test cases for a ticket, story, defect, or bug AND get them into their tracker — or mentions a work item ID together with testing, a Test Plan/Suite, test steps, expected/actual results, or QA verification. Trigger even on a bare "write tests for

framework-standardsskills/framework-standards/SKILL.md

The single source of truth for this plugin's automation-framework engineering standards — pull the right standard for a building block or a design decision, product-neutrally and grounded in official docs. Use whenever the kit (the tutor teaching a concept, the builder generating code, a reviewer checking it) or the QA directly needs "what is our standard for X", "how should I structure Y", "which test level fits this case", "POM vs raw selectors", "builder vs JSON test data", "how do we keep tests isolated", or "how do we reuse another module's functionality". It reads the reference files under `knowledge/standards/` and returns the relevant standard(s) — Purpose, When to use it (and when not), the official-doc link(s), a product-neutral code shape, and the anti-patterns. It does not generate framework code itself — it supplies the standard that the tutor teaches, the builder generates to, and the reviewer checks against.

kit-builderskills/kit-builder/SKILL.md

The orchestrator for the qa-engineering — build an API + UI test-automation framework brick by brick, running phases 0-5 as a state machine and pausing at every phase to say what it will do, recommend an option with its reason, act, verify, and report before moving on. Use whenever the user asks to "build a test framework brick by brick", "scaffold an automation framework step by step", "set up test automation one phase at a time", "run the test-automation kit", "continue the framework build", or is at any phase (foundation, module API suite, UI suite, ADO integration, CI) after phase 0's scan-and-confirm. It reads the confirmed flow map from `flow-map.json`, scaffolds the kit's product-agnostic `template/`, and delegates each capability to the kit's own native skills and guidance docs — it never reinvents what those already do.

MCP servers2

adostdio
command
npx
args
-y @azure-devops/[email protected] ${ADO_ORG} --authentication interactive
playwrightstdio
command
npx
args
-y @playwright/[email protected]

Plugin manifests1

plugins/qa-engineering/.claude-plugin/plugin.json
{
  "name": "qa-engineering",
  "version": "0.6.0",
  "types": "./types/index.d.ts",
  "description": "The single AI QA engineering plugin: one orchestrator for the whole loop — analyze a story or a module, author and approve test cases, build and generate API + UI automation, run the suite, and turn a failing run into a bug — human-gated at every tracker write. On first run it builds a frontend-first Knowledge Base of the product's features, flows and testable rules (discovered from the app's routes, deepened per module) that grounds story and module analysis. Publishing cases to a Test Plan/Suite and filing bugs support Azure DevOps today (other trackers planned); every other workflow is tracker-independent and works with no tracker at all. Also teaches and reviews a Playwright + TypeScript framework. Self-contained with its own native QA skills; ships no product data — the QA supplies tracker, API, auth and URLs locally through a git-ignored house-profile and .env, and the Knowledge Base is generated into the QA's own repo.",
  "author": {
    "name": "shironigun"
  },
  "keywords": [
    "testing",
    "automation",
    "playwright",
    "typescript",
    "zod",
    "api",
    "ui",
    "qa",
    "ado",
    "test-cases",
    "bug-filing",
    "orchestrator",
    "knowledge-base"
  ]
}

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[qa-engineering on Agent Plugins Marketplace](https://pluginsmp.com/plugins/qa-engineering)