Skip to main content
Glama
abhirise1981

AI Test Automation MCP Server

by abhirise1981
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Hybrid

🚀 Enterprise AI-Powered Multi-Agent Test Automation Platform

Playwright Tests Live Test Report TypeScript Playwright Appium Gatling Axe-Core pgvector

A state-of-the-art, enterprise-grade test automation platform unifying Web UI, Native Mobile (iOS/Android), REST/GraphQL APIs, Stripe Billing & Webhook Cryptographic Security, High-Concurrency Load Testing, WCAG 2.1 AA Accessibility Compliance, Vector Database Embeddings & NL-to-SQL (RAG), and an Autonomous Multi-Agent AI Layer with self-healing locators.

👉 📊 View Live Interactive Test Dashboard

🏛️ High-Level System Architecture

graph TB
    subgraph AI["🤖 AUTONOMOUS AI AGENTS LAYER — Pipeline Orchestrator (pipeline.ts)"]
        direction LR
        JIRA["1. Jira Agent<br/>Fetch Story + ACs"] -->|pipeline.ts| PLAN["2. Planner Agent<br/>Generate BRD"]
        PLAN -->|pipeline.ts| GEN["3. Generator Agent<br/>Write .spec.ts Code"]
        GEN -->|pipeline.ts| HEAL["4. Healer Agent<br/>Self-Heal Locators"]
    end

    subgraph MCP["🔌 MCP SERVER — Tool Orchestration"]
        direction LR
        T1["fetch_story"] ~~~ T2["generate_brd"] ~~~ T3["generate_tests"] ~~~ T4["heal_tests"] ~~~ T5["run_tests"]
    end

    subgraph EXEC["🧪 TEST EXECUTION LAYER"]
        direction LR
        WEB["🌐 Web UI E2E<br/>Playwright + POM"]
        APITST["🔌 REST + GraphQL<br/>APIRequestContext"]
        STRIPE["💳 Stripe Billing<br/>HMAC-SHA256"]
        A11Y["♿ Accessibility<br/>Axe-Core WCAG 2.1"]
        MOBILE["📱 Native Mobile<br/>Appium + WDIO"]
        PERF["⚡ Load Testing<br/>Gatling SDK"]
        NLSQL["🧠 NL-to-SQL<br/>Vector RAG"]
    end

    subgraph INFRA["☁️ CI/CD & INFRASTRUCTURE LAYER"]
        direction LR
        LOCAL["Chromium / Firefox / WebKit"]
        BS["BrowserStack Cloud<br/>Real iOS + Android"]
        CICD["GitHub Actions + GitLab CI"]
        RPT["📊 Live Dashboard<br/>GitHub Pages"]
    end

    AI --> MCP
    MCP --> EXEC
    EXEC --> INFRA

Related MCP server: Vibe Testing

💎 Framework Modules & Key Capabilities

Layer

Technologies

Highlights & Test Coverage

🌐 Web UI E2E

Playwright, TypeScript, Page Object Model (POM)

Full customer journeys: 2-step Signup/Login, Product Discovery, Cart management, and Checkout order placement. Resilient against ad intercepts and dynamic modal animations.

💳 Stripe & Billing

TypeScript, HMAC-SHA256, Form-URL-Encoded REST

End-to-end SaaS subscription lifecycle (Starter, Pro, Enterprise), PaymentIntents, partial/full refunds, webhook replay protection, and cryptographic signature validation.

📱 Native Mobile

Appium, WebdriverIO, Screen Object Model (SOM)

Cross-platform automation on iOS Simulator and Android Emulator. Touch gestures, scrolls, and device orientation handling.

📱 Mobile-Web

Playwright Device Descriptors

Mobile viewport emulation (iPhone 14, Pixel 7) testing responsive hamburger navigation and touch tap targets.

🔌 REST & GraphQL

Playwright APIRequestContext, Service Object Model

Full CRUD assertions (GET, POST, PUT, PATCH, DELETE), Basic Auth security boundaries, reverse geocoding, and GraphQL query resolution.

⚡ Performance

Gatling TypeScript SDK, P.S.I.A. Architecture

Concurrency load tests, token correlation, cookie header management, and response time SLA assertions.

♿ Accessibility (a11y)

Axe-Core, WCAG 2.1 AA, Section 508

Automated WCAG compliance auditing, keyboard navigation verification (Tab focus order), 4.5:1 color contrast enforcement, 48×48px touch target validation, and component-scoped modal scanning with third-party widget exclusion.

🧠 Vector DB & NL-to-SQL

Embeddings, Cosine Similarity, pgvector, RAG

High-dimensional vector embeddings for semantic schema retrieval (Recall@K), NL-to-SQL query generation with SQL injection guardrails, and intelligent test deduplication via cosine similarity.

🤖 AI Self-Healing

LangChain, LLM Cache, AST Repair

Autonomous test generation from Jira BRDs + self-healing locator engine that heals broken UI selectors at runtime via vector similarity matching.

🚀 Quickstart & Execution Guide

1. Installation

# Clone the repository
git clone https://github.com/abhirise1981/ai-test-automation-platform.git
cd ai-test-automation-platform

# Install dependencies
npm install

# Install Playwright browser binaries
npx playwright install --with-deps

2. Run Test Suites

# Run all automated tests (UI, REST, GraphQL, Stripe, Accessibility, AI)
npm run test

# Run UI E2E tests only
npx playwright test tests/ui/

# Run Stripe Billing & Webhook tests only
npx playwright test tests/api/stripe.spec.ts

# Run REST & GraphQL API tests only
npx playwright test tests/api/location.spec.ts tests/api/graphql.spec.ts

# Run Accessibility (WCAG 2.1 AA) tests
npx playwright test tests/ui/accessibility.spec.ts

# Run Vector DB & NL-to-SQL (RAG) tests
npx playwright test tests/ai/

# Run Cross-Browser tests (Chromium + Firefox + WebKit)
npm run test:cross-browser

# Run Mobile-Web responsive tests
npm run test:mobile-web

3. Native Mobile Automation (Appium)

# Start Appium Server
npm run appium:start

# Run Android Native Tests
npm run test:mobile:android

# Run iOS Native Tests
npm run test:mobile:ios

# Run on BrowserStack Real Devices
npm run test:mobile:browserstack

4. Gatling Performance & Load Testing

# Run Gatling TypeScript Load Tests
npx gatling run --typescript --sources-folder load-tests --simulation ecommerce

5. AI Multi-Agent Pipeline & MCP Server

# Run the autonomous Jira-to-Test agent pipeline
npx ts-node agents/pipeline.ts --issue PROJ-123

# Start the MCP Tool Server for AI assistants
npm run mcp:start

♿ Accessibility Testing (WCAG 2.1 AA)

The framework includes a purpose-built A11yAuditor engine (utils/A11yAuditor.ts) powered by Axe-Core for automated WCAG 2.1 AA compliance:

Capability

WCAG Standard

Implementation

Full Page Auditing

WCAG 2.0/2.1 Level A & AA, Section 508

Tag-based Axe-Core rulesets with severity gating

Keyboard Navigation

WCAG 2.1.1 & 2.4.7

Sequential Tab focus order assertions

Color Contrast

WCAG 1.4.3

≥ 4.5:1 ratio enforcement for normal text

Touch Target Size

WCAG 2.5.5 / 2.5.8

≥ 48×48px bounding box validation

Component Scoping

—

Isolate modals/dialogs, exclude third-party ad widgets

Severity Quality Gate

—

CI build fails on critical or serious violations

# Run Accessibility tests
npx playwright test tests/ui/accessibility.spec.ts

🧠 Vector Database, Embeddings & NL-to-SQL (RAG)

The platform includes a Vector Store (agents/vector/VectorStore.ts) and NL-to-SQL Agent (agents/sql/NlToSqlAgent.ts) for AI-powered schema retrieval and natural language query generation:

graph LR
    NL["Natural Language Assertion"] --> EMB["Embedding Vector"]
    EMB --> VDB[("Vector Store / pgvector")]
    VDB --> SIM["Cosine Similarity Search"]
    SIM --> RAG["Top-K Schema Retrieval (RAG)"]
    RAG --> SQL["Safe Parameterized SQL"]
    SQL --> GUARD["SQL Injection Guardrails"]

Capability

What It Does

Embedding Generation

Converts text into high-dimensional normalized vectors using character n-gram frequency hashing

Cosine Similarity Search

k-NN semantic search for retrieving the most relevant schemas

Schema RAG (Retrieval-Augmented Generation)

Indexes table schemas into vector store; retrieves only relevant tables for SQL generation

NL-to-SQL Translation

Translates natural language requirements into verified SQL queries

SQL Safety Guardrails

Rejects destructive DDL/DML (DROP, DELETE, TRUNCATE, ALTER, ; --)

Test Deduplication

Identifies semantically duplicate test cases via similarity thresholds

Self-Healing Locators

Matches broken CSS/XPath selectors to new DOM elements by semantic meaning

# Run Vector DB & NL-to-SQL tests
npx playwright test tests/ai/nl-to-sql-vector.spec.ts

📊 Reports & CI/CD Pipelines

  • Dual CI/CD Integration:

    • GitHub Actions: Automated regression on every push/PR with instant artifact publishing (Node.js 22 LTS, 2 parallel CI workers, npm ci).

    • GitLab CI: Fully configured .gitlab-ci.yml multi-stage pipeline (Build → Test → Performance).

  • Interactive Live Reporting:

  • Local Report Viewer:

    npx playwright show-report
# or Monocart Executive Dashboard:
npx monocart show-report test-results/report.html

📄 License

Distributed under the MIT License. Developed for enterprise-grade test automation and architectural evaluations.

This server cannot be installed

F
license - not found
Not graded
quality - not tested
B
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • BugEzy

    Voice-powered bug reporting with 13 MCP tools. Record bugs by talking; let AI find and fix them.

  • mcp

    Screenshot, diff, audit and sitemap-capture any web page — 5 MCP tools for AI agents.

  • Frenchie

    OCR, transcription, file extraction, and image generation for AI agents via MCP.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/abhirise1981/ai-test-automation-platform'

If you have feedback or need assistance with the MCP directory API, please join our Discord server