QA AI Toolkit

A Cursor plugin for QA, testing, debugging, code review, verification workflows, API artifact generation, UI testing, Excel test case generation, static analysis, and test planning.

InitQA — Full Pipeline Orchestration

The initqa skill runs the entire QA pipeline in one command. It detects your tech stack and executes every relevant skill in the correct order:

When to use it:

• Onboarding a new project to qa-ai-toolkit
• Sprint kickoff — generate all test artifacts before development begins
• Setting up CI/CD test infrastructure from scratch
• Audit — verify all test layers exist for an existing codebase

Supported stacks: TypeScript / Node.js / NestJS / Express, Java / Spring Boot / Quarkus, Python / FastAPI / Django / Flask, Go / Gin / Echo / Fiber, C# / ASP.NET Core, Rust / Actix / Axum — Angular, React, Vue, Svelte, Next.js, Nuxt.js, SvelteKit — Nx / Turborepo / Lerna / pnpm workspaces

What's Included

Pipeline Orchestrator (1)

| Skill | Triggers When | What It Does | |-------|---------------|--------------| | initqa | User wants to run the full QA pipeline end-to-end, onboard a new project, or set up complete test coverage in one run | Detects tech stack (frontend/backend/framework), then orchestrates all qa-ai-toolkit skills in the correct order across 6 phases: detection, run-mode selection, test planning, backend track, frontend track, code review, and verification gate |

Skills (17)

| Skill | Triggers When | What It Does | |-------|---------------|--------------| | fsd-test-case-generator | Generating test cases from requirement documents (FSD, PRD, BRS) in a folder | Reads requirement documents, extracts features and user stories, builds a traceability matrix, and generates a multi-sheet Excel workbook with test cases — manual-only, never auto-invoked by initqa | | test-plan-generator | Analyzing a feature, PR, or requirements doc before writing tests | Produces structured test plan with prioritized scenarios (P0-P3), test types, edge cases, and execution recommendations | | postman-collection-generator | Creating Postman collections for manual API testing | Generates organized Postman Collection v2.1 with environment files, variables, and pre-request scripts | | postman-test-suite-generator | Creating API test coverage with positive and negative scenarios | Produces Postman test collection with systematic positive and negative test cases for every endpoint | | newman-runner | Automating Postman test execution in CI/CD | Generates Newman CLI config, npm scripts, and CI/CD pipeline configs (GitHub Actions, GitLab CI, Jenkins, Azure DevOps, Bitbucket Pipelines) for automated API test execution | | backend-excel-test-generator | Generating function-level test cases from backend source | Generates Excel (.xlsx) test cases from TypeScript, JavaScript, Python, Java, Node.js, .NET (C#), or Go backend source by analyzing functions, classes, API handlers, services, and repositories | | vitest-test-generator | Source code needs unit or integration test coverage | Generates Vitest/Jest test files with describe/it blocks, mocking, edge cases, and async testing patterns | | unit-test-generator | .NET (C#) or Java source needs unit test coverage | Detects language, test framework (xUnit/NUnit/MSTest or JUnit 5/TestNG), and mocking library (Moq/NSubstitute or Mockito/AssertJ); generates test files with AAA pattern, mocking, edge cases, and project config — standalone, not part of any pipeline | | playwright-test-generator | Generating E2E tests for Angular, React, Vue, or Svelte frontends | Auto-detects framework, scans components for interactions, generates Playwright tests with Page Object Model covering forms, clicks, navigation, modals, accessibility, and responsive behavior | | web-excel-test-generator | Generating function-level test cases from frontend source | Generates Excel (.xlsx) test cases for Angular, React, or Vue projects by analyzing components, services, hooks, composables, and utility functions | | uiux-navigation-test-generator | Generating navigation flow test cases from routing config | Generates Excel (.xlsx) navigation flow test cases and Playwright/Appium test stubs by analyzing routing configs, navigation trees, and screen transitions | | code-review-debug-analyzer | Performing static code review without running the application | Extracts business logic, detects design patterns, identifies bugs and security gaps; outputs comprehensive Excel findings sheet for developer self-verification before QA handoff | | requesting-code-review | Completing tasks or before merging | Dispatches code-reviewer agent with structured checklist and git diff range | | receiving-code-review | Receiving code review feedback | Ensures technical evaluation over performative agreement; verify before implementing | | systematic-debugging | Encountering any bug, test failure, or unexpected behavior | 4-phase process: root cause investigation, pattern analysis, hypothesis testing, implementation | | verification-before-completion | About to claim work is complete or passing | Requires running verification commands and confirming output before making success claims | | git-diff-updater | Syncing QA artifacts with recent git changes (manual-only) | Diffs git range and updates only affected QA artifacts (Postman, Excel, Playwright, Vitest) without regenerating the full pipeline; preserves Dev Verified status |

Agent (1)

| Agent | Purpose | |-------|---------| | code-reviewer | Senior code reviewer that checks plan alignment, code quality, architecture, testing, and production readiness |

Review Prompt Templates (2)

| Template | Purpose | |----------|---------| | spec-reviewer-prompt | Verify implementation matches specification (nothing more, nothing less) | | code-quality-reviewer-prompt | Verify implementation is well-built (clean, tested, maintainable) |

API Artifact Generation

The API skills form a pipeline that takes your codebase or API design document and produces ready-to-use testing artifacts with automated CI/CD execution.

Workflow:

Codebase / OpenAPI Spec
  → postman-collection-generator  (requests, folders, env vars, pre-request scripts)
  → postman-test-suite-generator  (positive + negative test cases per endpoint)
  → newman-runner                 (CLI config, npm scripts, CI/CD pipelines)

1. Generate the Postman collection — builds organized requests with folders, environment variables, realistic example data, and pre-request scripts from codebase analysis or an existing OpenAPI spec
2. Generate the test suite — adds comprehensive positive and negative test scenarios covering missing fields, invalid types, boundary values, auth failures, not-found cases, duplicates, and more
3. Automate execution — generates Newman CLI configuration, npm scripts, and CI/CD pipeline configs for automated test execution on every push

Each skill can also run independently. The Postman collection generator can work directly from codebase analysis, and the test suite generator can work from an OpenAPI spec.

UI Test Generation

The Playwright skill handles frontend UI test generation for Angular, React, Vue, and Svelte applications:

What it covers:

• Auto-detection — identifies Angular, React, Vue, or Svelte; component libraries (Material, MUI, Ant Design, Vuetify, Element Plus); and routing configuration
• Component scanning — discovers click handlers, form bindings, modals, dropdowns, tables, navigation links, conditional rendering
• Positive tests — page load, valid form submission, navigation, modal interactions, table sorting and pagination
• Negative tests — empty required fields, invalid formats, boundary values, unauthorized access, empty/error states, double-submit prevention
• Responsive tests — mobile hamburger menu, sidebar collapse, stacked layouts
• Accessibility tests — keyboard navigation, ARIA labels, focus management, axe-core audits

Excel Test Case Generation

Four skills generate structured Excel (.xlsx) test case workbooks for different layers of the application:

| Skill | Input | Output | |-------|-------|--------| | fsd-test-case-generator | Requirement documents (FSD, PRD, BRS, markdown, images) | Multi-sheet workbook with traceability matrix, features, user stories, and test cases | | backend-excel-test-generator | Backend source (TS, JS, Python, Java, C#, Go) | Function-level test cases from API handlers, services, repositories, utilities | | web-excel-test-generator | Frontend source (Angular, React, Vue) | Function-level test cases from components, services, hooks, composables | | uiux-navigation-test-generator | Routing config and navigation trees | Navigation flow test cases + Playwright/Appium test stubs |

Each workbook includes positive/negative scenarios, boundary values, expected results, and priority levels — ready for developer self-verification before QA handoff.

Static Analysis & Code Review

The code-review-debug-analyzer skill performs deep static analysis without running the application:

Source Code (fullstack)
  → Extract business logic and control flow
  → Detect design patterns and strategies
  → Identify bugs, code smells, and security gaps
  → Output: Excel findings workbook for developer verification

Supported stacks: TypeScript, JavaScript, Python, Java, C#, Go backends; React, Angular, Vue, Next.js, Svelte frontends; React Native, Flutter, iOS Swift, Android Kotlin mobile

Unit & Integration Testing

Two skills cover unit and integration test generation:

vitest-test-generator (TypeScript / JavaScript)

• Framework detection — auto-detects Vitest, Jest, or existing test setup
• Code analysis — scans functions, classes, API handlers, and React/Angular components
• Test generation — describe/it blocks with happy path, edge case, and error scenarios
• Mock setup — external dependencies mocked with proper cleanup patterns

unit-test-generator (.NET / Java — standalone)

• Language detection — C# (.csproj) or Java (pom.xml / build.gradle)
• Framework detection — xUnit, NUnit, MSTest for .NET; JUnit 5, TestNG for Java
• Mock library detection — Moq, NSubstitute for .NET; Mockito, AssertJ for Java
• Test generation — AAA pattern, constructor injection mocking, edge cases, async support

Test Planning

The test-plan-generator skill produces structured test plans before writing any test code:

• Input sources — feature descriptions, PR diffs, requirements documents, user stories
• Scenario coverage — positive, negative, edge case, security, and performance scenarios
• Priority levels — P0 (blocks release) through P3 (nice to have)
• Execution mapping — recommends which qa-ai-toolkit skills to use for each scenario type

Git Diff Updater (Manual-Only)

The git-diff-updater skill syncs QA artifacts with recent code changes without regenerating the full pipeline:

Git Diff (commit range / uncommitted changes)
  → Identify affected files and functions
  → Update only impacted QA artifacts (Postman, Excel, Playwright, Vitest)
  → Preserve Dev Verified status on unchanged test cases

This skill is never auto-invoked — use it explicitly when you need incremental artifact updates after code changes.

Installation

Via npm (recommended)

npx qa-ai-toolkit install

Via script

# One-liner (after hosting the repo)
curl -sSL https://raw.githubusercontent.com/<your-org>/qa-ai-toolkit/main/install.sh | bash

# Or clone and run the script
git clone <repo-url> && cd qa-ai-toolkit && ./install.sh

See INSTALLATION.md for the full guide — npm CLI commands, manual install, environment variables, update/uninstall commands, team distribution options, and troubleshooting.

Usage

Skills activate automatically based on context. You can also invoke them explicitly:

Use qa-ai-toolkit:initqa
Use qa-ai-toolkit:fsd-test-case-generator
Use qa-ai-toolkit:test-plan-generator
Use qa-ai-toolkit:postman-collection-generator
Use qa-ai-toolkit:postman-test-suite-generator
Use qa-ai-toolkit:newman-runner
Use qa-ai-toolkit:backend-excel-test-generator
Use qa-ai-toolkit:vitest-test-generator
Use qa-ai-toolkit:unit-test-generator
Use qa-ai-toolkit:playwright-test-generator
Use qa-ai-toolkit:web-excel-test-generator
Use qa-ai-toolkit:uiux-navigation-test-generator
Use qa-ai-toolkit:code-review-debug-analyzer
Use qa-ai-toolkit:requesting-code-review
Use qa-ai-toolkit:receiving-code-review
Use qa-ai-toolkit:systematic-debugging
Use qa-ai-toolkit:verification-before-completion
Use qa-ai-toolkit:git-diff-updater

License

MIT