K11 Tech Lab – Selenium Java Automation Framework

Getting Started with the Java-Based End-to-End Test Automation Framework:

This Java-based test automation framework is designed to support end-to-end test coverage across web, API, and database layers, with seamless CI/CD integration and enterprise-grade extensibility.

---

✅ Core Features

• Web UI Testing using Selenium WebDriver
• API Testing with REST-assured and Java SAAJ for SOAP
• Database Validation via JDBC with SQL/NoSQL utility support
• Test Data Management with Excel, JSON, database queries, and runtime data generation
• Dynamic Configuration using .properties, .yaml, or .json with CLI/CI parameters
• Structured Logging & Reporting with Log4j/SLF4J, ExtentReports, email and screenshot support
• CI/CD Ready: Jenkins, GitHub Actions, GitLab, and cloud execution (e.g., Sauce Labs)

🚀 AI-Powered Features

• Self-Healing Test Automation:
  • AI-powered element locator healing for resilient UI tests
  • Automatically recovers from locator changes using semantic context and RAG
  • Reduces test flakiness and maintenance effort
• Retrieval-Augmented Generation (RAG):
  • Semantic search and answer synthesis over your test docs and codebase
  • Supports OpenAI, HuggingFace, and local Ollama embeddings
  • Multi-provider, local/cloud, and offline AI support
• Persistent Embedding Cache:
  • Embeddings are computed once per document chunk and reused for all future runs
  • Massive speedup and cost savings for repeated queries
• Chatbot/NLP & Conversational AI:
  • Conversational UI for natural language queries
  • Integrates with RAG and knowledge base for context-aware answers
  • Supports both local and cloud LLMs

🧠 Model Control Plane (MCP) Server — AI-Driven Test Generation & Review

• Lightweight Java HTTP server for AI-powered test automation workflows
• Endpoints for:
  • Page object generation from prompt files
  • Playwright/Selenium test generation and execution
  • Automated code review of generated tests (/mcp/test-code-review)
  • Reporting: lists test run artifacts and logs (/mcp/report)
• Pluggable AI clients: Supports RAG, OpenAI, and local LLMs for prompt completion and code review
• Playwright bridge: Runs Playwright tests via a Node.js subprocess for browser automation
• MongoDB context store: (optional) for storing workflow and conversation context
• Demo/test script: scripts/test_mcp.sh exercises all endpoints for quick validation

Quick start:

# Start the MCP server (from project root)

# Or run directly with Java if built:


# In another terminal, run the demo script:
# 🚀 AI-Powered Features

**Endpoints:**
- `POST /mcp/generate-page-object` — Generate page objects from prompt
- `POST /mcp/generate-and-run-playwright-test` — Generate & run Playwright test
- `POST /mcp/generate-and-run-selenium-test` — Generate & run Selenium test
- `POST /mcp/test-code-review` — Automated AI code review for test code
- `GET  /mcp/report` — List test run reports and logs

**Usage Example:**

Start the MCP server:
```bash
mvn exec:java -Dexec.mainClass="org.k11techlab.framework.ai.mcp.MCPServer"

Generate a page object from a prompt file:

curl -X POST http://localhost:8090/mcp/generate-page-object \
  -H "Content-Type: application/json" \
  -d '{"promptFile":"k11softwaresolutions/pages/pageobject_creation_prompt_multi.txt"}'

Review a test code snippet with AI:

curl -X POST http://localhost:8090/mcp/test-code-review \
  -H "Content-Type: text/plain" \
  --data-binary @YourTestFile.java

Fetch a list of test run reports and logs:

curl http://localhost:8090/mcp/report

Built entirely with open-source libraries, this framework is fully extensible—ready to scale for validations involving files, emails, microservices, or third-party system integrations.

---

🧱 Architecture Overview

📊 High-Level Diagram

Selenium Automation Framework Architecture — © 2025 Kavita Jadhav. All rights reserved.

---

🤖 AI Features Integration Architecture

flowchart TD
   A[Test Suite] --> B(Core Framework)
   B --> C[AI Integration Layer]
   C --> D[RAG Engine]
   C --> E[Self-Healing Engine]
   C --> J[Chatbot/NLP & Conversational AI]
   C --> M[MCP Server]
   M --> C
   M --> N[Playwright/Selenium Bridge]
   M --> O[Prompt Files]
   D --> F[Embedding Providers]
   D --> G[Embedding Cache]
   D --> H[Docs/Knowledge Base]
   E --> I[Locator Healing]
   J --> K[Conversational UI]

Loading

AI Features Integration: Shows how RAG, embedding cache, and self-healing plug into the core automation framework. Render this diagram with Mermaid for a visual overview.

🤖 AI Enhancements (2025)

This framework now includes advanced AI-powered features for smarter, context-aware automation and documentation search:

• Self-Healing Test Automation:
  • AI-powered element locator healing for resilient UI tests
  • Automatically recovers from locator changes using semantic context and RAG
  • Reduces test flakiness and maintenance effort
• Retrieval-Augmented Generation (RAG):
  • Semantic search and answer synthesis over your test docs and codebase
  • Supports OpenAI, HuggingFace, and local Ollama embeddings
  • Multi-provider, local/cloud, and offline AI support
• Persistent Embedding Cache:
  • Embeddings are computed once per document chunk and reused for all future runs
  • Massive speedup and cost savings for repeated queries
• Model Control Plane (MCP) Server:
  • Lightweight Java HTTP server for AI-powered test generation, review, and workflow orchestration
  • Exposes endpoints for page object generation, Playwright/Selenium test execution, automated code review, and reporting
  • Integrates with RAG, OpenAI, and local LLMs; bridges to Playwright/Selenium; supports prompt-driven automation
• AI Demo & Documentation:
  • SelfHealingDemoTest.java (Unit Test)
  • RAGComponentsDemo.java (RAG Test Class)
  • Sample Output (Cache Hits!)
  • RAG Demo Article & How-To

These enhancements make the framework ready for next-generation, AI-assisted, and self-healing test automation and knowledge retrieval.

The framework is composed of well-structured layers to ensure modularity, maintainability, and scalability across complex enterprise test environments.

1️⃣ Framework Layer – The Foundation

• Driver management (Selenium Grid/local/cloud)
• Config loading from external files
• Page Object Model (POM) structure
• Test data provider (Excel/JSON/DB)

2️⃣ Utility Classes – Powering Reusability

• Wait utilities (explicit/implicit/fluent)
• File, JSON, Excel handlers
• REST & SOAP service clients
• DB interaction (JDBC-based)
• Locator and email utilities

3️⃣ Automated Test Suite – The Execution Brain

• Test cases built on Base Test structure
• POM-based interactions
• Data-driven via @DataProvider
• Configurable execution (env, role, browser)
• Domain-Specific Language (DSL) support for readability

4️⃣ Test Execution – Anywhere, Anytime

• Run tests locally, via Docker, VMs, or cloud (Sauce Labs, BrowserStack)
• Supports headless execution
• Retry analyzer and failure recovery
• Data cleanup & environment reset utilities

5️⃣ CI/CD Integration – Automating the Pipeline

• Jenkins / GitHub Actions ready
• Parameterized build support
• Maven-based dependency management
• Artifactory/Nexus for internal libs

6️⃣ Error Handling and Recovery Scenarios – Resilience First

• Centralized exception handling
• Custom exception types
• Retry mechanism (TestNG-based)
• Safe teardown and recovery logic

7️⃣ Logging and Reporting – Know What Happened, Instantly

• ExtentReports/Allure HTML reports
• Log4j/SLF4J structured logs
• Screenshot capture on failure
• Email notifications with test summaries

8️⃣ Framework Capabilities & Extensibility

• Supports Web, Mobile, SOAP, REST API testing
• Dynamic configuration & data handling
• Cloud-ready & DevOps integrated
• Extensible for:
  • File-based validations (local/FTP)
  • Email workflows
  • Microservices architecture
  • Localization, accessibility, performance testing

---

🏗️ Technologies Used

• Java 8+
• Selenium WebDriver
• REST-assured
• SAAJ API
• TestNG
• Apache POI / Jackson / Gson
• Log4j / SLF4J
• ExtentReports / Allure
• JDBC
• Maven

---

🔧 Setup & Configuration

📦 Prerequisites

• Java 11+
• Maven 3.6+
• Git
• Chrome or Firefox browser
• IDE (e.g., IntelliJ, Eclipse)

🚀 Clone the Repository

git clone https://github.com/K11-Software-Solutions/k11-techlab-selenium-java-automation-framework.git
cd k11-techlab-selenium-java-automation-framework

⚙️ Configuration

1. Edit config files in:

   src/test/resources/config/
   

   Customize:

   • baseUrl
   • browser
   • Timeouts, credentials, etc.

2. Ensure browser drivers (e.g., ChromeDriver) are in system path or configured in test base.

3. Use TestNG XML for specific suites:

   src/test/resources/testng/
   

---

🧪 Running Tests

mvn clean test
mvn clean test -DsuiteXmlFile=smoke.xml

---

📊 Reporting

mvn allure:report
allure serve target/allure-results

---

📦 Use Cases

• UI regression and smoke testing
• Cross-browser automation
• Framework learning or extension baseline
• CI integration with test reporting

---

📄 License

This project is licensed under the MIT License – see the LICENSE file for details.

---

📬 Contact

For consulting, training, or implementation support:
🔗 softwaretestautomation.org
🔗 k11softwaresolutions.com
📧 k11softwaresolutions@outlook.com