diff --git a/.cursor/plugin.json b/.cursor/plugin.json
new file mode 100644
index 0000000..c2e23d8
--- /dev/null
+++ b/.cursor/plugin.json
@@ -0,0 +1,24 @@
+{
+  "name": "browserstack",
+  "version": "1.0.0",
+  "description": "BrowserStack integration for Cursor. Test websites and mobile apps on real devices, run automated tests, debug failures, and manage test cases—all using natural language.",
+  "author": {
+    "name": "BrowserStack",
+    "email": "support@browserstack.com",
+    "url": "https://www.browserstack.com"
+  },
+  "keywords": [
+    "browserstack",
+    "testing",
+    "automation",
+    "accessibility",
+    "quality-assurance",
+    "playwright",
+    "selenium",
+    "manual-testing",
+    "app-testing",
+    "real-devices"
+  ],
+  "logo": "https://github.com/browserstack/mcp-server/raw/main/assets/browserstack-logo.png",
+  "primaryColor": "#0070f0"
+}
diff --git a/README.md b/README.md
index 36d0340..c95dfa8 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,450 @@
-# browserstack-cursor-plugin
-BrowserStack Cursor Plugin
+# BrowserStack for Cursor
+
+Test websites and mobile apps on 3,500+ real devices and browsers—directly from Cursor using natural language. No context switching and very easy to setup.
+
+## What You Can Do
+
+**Test anywhere, from anywhere**
+- Open your localhost app on an iPhone 15 Pro Max
+- Test your website on Safari 17 without owning a Mac
+- Debug app crashes on Android 14 in real-time
+- Run Playwright tests on 50+ browser/OS combinations
+
+**Stay in your flow**
+- Ask in plain English: *"Test my site on Edge"*
+- Get instant access to real devices
+- Debug failed tests without leaving Cursor
+- Fix accessibility issues with AI suggestions
+
+**Automate everything**
+- Run test suites on BrowserStack infrastructure
+- Generate test cases from product requirements
+- Auto-heal flaky tests with AI
+- Manage test cases and track results
+
+## Setup
+
+### Step 1: Install Node.js
+
+You need Node.js version 18 or higher (we recommend v22.15.0 LTS).
+
+Check your version:
+```bash
+node --version
+```
+
+If you need to install or update:
+- **macOS**: `brew update && brew upgrade node` or [download here](https://nodejs.org/en/download)
+- **Windows**: Download from [nodejs.org](https://nodejs.org/en/download)
+
+### Step 2: Get BrowserStack Credentials
+
+1. **Sign up** at [browserstack.com/users/sign_up](https://www.browserstack.com/users/sign_up)
+   - Free trial available, no credit card required
+   - Open source projects get free access
+
+2. **Get your credentials** from [Account Settings](https://www.browserstack.com/accounts/profile/details)
+   - You'll need your `Username` and `Access Key`
+   - Keep these handy for the next step
+
+### Step 3: Install the Plugin
+
+1. **Install from Cursor Marketplace** (or manually)
+
+2. **Add your credentials** to your environment:
+
+   **On macOS/Linux:**
+   ```bash
+   # Open your shell profile
+   nano ~/.zshrc  # or ~/.bashrc if you use bash
+   
+   # Add these lines at the end (replace with your actual credentials)
+   export BROWSERSTACK_USERNAME="your_username_here"
+   export BROWSERSTACK_ACCESS_KEY="your_access_key_here"
+   
+   # Save and reload
+   source ~/.zshrc
+   ```
+
+   **On Windows (PowerShell):**
+   ```powershell
+   # Open PowerShell profile
+   notepad $PROFILE
+   
+   # Add these lines
+   $env:BROWSERSTACK_USERNAME="your_username_here"
+   $env:BROWSERSTACK_ACCESS_KEY="your_access_key_here"
+   ```
+
+3. **Restart Cursor** to activate the plugin
+
+4. **Verify installation** by asking Cursor:
+   ```
+   "Open google.com on Chrome"
+   ```
+
+✅ That's it! You're ready to test.
+
+## Common Workflows
+
+### 🚀 Quick Manual Testing
+
+Testing your local development site on different browsers:
+
+```
+💬 "Open my website running on localhost:3000 on Safari 17"
+💬 "Test localhost:8080 on the latest Chrome on Windows 11"
+💬 "Open my site on Firefox on macOS"
+💬 "Take a screenshot of this page"
+```
+
+**Use case**: You're developing a feature and need to check if it works on Safari, but you're on Windows.
+
+**Pro tip**: If you find a visual bug, use the **Visual Issue Fix** skill to get browser-specific fixes.
+
+---
+
+### 📱 Mobile App Testing
+
+Test your mobile app on real devices:
+
+```
+💬 "Open my app on iPhone 15 Pro Max"
+   (Cursor will ask for your .ipa or .apk file path)
+
+💬 "Test my app on Galaxy S24 with Android 14"
+💬 "My app crashes on iOS 17 when I tap the login button, help me debug"
+```
+
+**Use case**: You don't have physical access to an iPhone 15 Pro Max but need to verify the login flow works.
+
+---
+
+### 🧪 Running Automated Tests
+
+Run your existing test suite on BrowserStack:
+
+```
+💬 "Setup my Playwright tests to run on BrowserStack"
+💬 "Run my tests on Chrome, Firefox, and Safari"
+💬 "My test suite failed on BrowserStack session abc123, help me debug"
+```
+
+**Workflow example**:
+1. You: *"Setup my Playwright tests to run on BrowserStack"*
+2. Cursor updates your config files automatically
+3. You: *"Run my tests on Chrome and Edge"*
+4. BrowserStack runs your tests, Cursor shows you results
+5. If tests fail: *"Help me debug the failures"*
+
+---
+
+### ♿ Accessibility Scanning
+
+Catch accessibility issues before production (uses **Accessibility Fix** skill):
+
+```
+💬 "Scan accessibility issues on localhost:3000"
+💬 "Fix accessibility problems on my checkout page"
+💬 "Check WCAG 2.1 AA compliance for my signup form"
+```
+
+**Use case**: You want to ensure your signup form is accessible before deploying.
+
+**What happens**: The skill scans your page, identifies violations, provides specific code fixes, and verifies after you implement them.
+
+---
+
+### 📋 Test Case Management
+
+Organize and track your testing:
+
+```
+💬 "Create a Test Management project called 'Payment Flow'"
+💬 "Add a test case: Verify credit card payment with invalid CVV"
+💬 "List all high priority test cases in Payment Flow"
+💬 "Create a test run for smoke tests"
+💬 "Mark test case TC-123 as passed"
+```
+
+**Workflow example**:
+1. Create project structure
+2. Add test cases as you develop features
+3. Create test runs before releases
+4. Track results and generate reports
+
+---
+
+### 🤖 AI-Powered Testing
+
+**Generate test cases from requirements** (uses **Test Case Generator** agent):
+```
+💬 @test-case-generator
+   [Attach your PRD file]
+   "Generate test cases for project PR-12345"
+   
+💬 @test-case-generator
+   "Create test cases for checkout flow with credit card and PayPal"
+   "Add to project PR-67890"
+```
+
+**Auto-heal flaky tests:**
+```
+💬 "My login test keeps failing because the button selector changed. Fix it with self-healing."
+💬 "Get self-healed selectors for session session_abc123"
+```
+
+**Convert manual tests to automation:**
+```
+💬 "Convert my manual test case TC-45 into a Playwright script"
+```
+
+---
+
+## Built-in Skills & Agents
+
+The plugin includes specialized workflows and AI agents to streamline common testing tasks.
+
+### 🔧 Skills (Multi-Step Workflows)
+
+#### Accessibility Fix
+**What it does**: Scans your page for WCAG violations, identifies issues, and provides code fixes.
+
+**When to use**: Before deployment, fixing accessibility bugs, ensuring compliance.
+
+**Sample commands**:
+```
+💬 "Scan accessibility issues on localhost:3000"
+💬 "Fix the accessibility problems on my checkout page"
+💬 "Check WCAG 2.1 AA compliance for localhost:8080/signup"
+```
+
+**Workflow**:
+1. Runs accessibility scan on your page
+2. Categorizes issues (Critical/High/Medium)
+3. Provides specific code fixes for each issue
+4. Re-scans after you implement fixes to verify
+
+**Example fix output**:
+```html
+<!-- Missing form label fix -->
+<!-- Before -->
+<input type="email" placeholder="Email">
+
+<!-- After -->
+<label for="email">Email Address</label>
+<input type="email" id="email" aria-required="true">
+```
+
+---
+
+#### Visual Issue Fix
+**What it does**: Captures screenshots across browsers, identifies visual bugs, and provides CSS fixes.
+
+**When to use**: Cross-browser layout issues, responsive design bugs, visual regressions.
+
+**Sample commands**:
+```
+💬 "My navigation is broken in Safari, help me fix it"
+💬 "Compare how my homepage looks on Chrome vs Firefox"
+💬 "Fix the layout issue on Safari for localhost:3000"
+```
+
+**Workflow**:
+1. Opens page on target browsers
+2. Captures screenshots for comparison
+3. Identifies visual differences
+4. Provides browser-specific CSS fixes
+5. Validates fix doesn't break other browsers
+
+**Example fix output**:
+```css
+/* Safari flexbox gap issue fix */
+/* Before */
+.container {
+  display: flex;
+  gap: 20px;
+}
+
+/* After - Works everywhere */
+.container {
+  display: flex;
+}
+.container > * {
+  margin-right: 20px;
+}
+```
+
+---
+
+### 🤖 AI Agent
+
+#### Test Case Generator
+**What it does**: Generates comprehensive test cases from requirements, user stories, or PRD documents using BrowserStack AI.
+
+**When to use**: Planning new features, creating test suites, converting requirements to test cases.
+
+**Sample commands**:
+```
+💬 @test-case-generator
+   [Add your PRD file or requirements document to context]
+   "Generate test cases for Test Management project PR-12345"
+
+💬 @test-case-generator
+   "Create test cases for a login feature with email and Google OAuth"
+   "Add them to project PR-54321"
+```
+
+**Workflow**:
+1. Invoke agent with `@test-case-generator`
+2. Add requirements file to context (or describe feature)
+3. Provide your Test Management project ID (format: `PR-12345`)
+4. Agent generates structured test cases
+5. Test cases are added to your BrowserStack Test Management project
+
+**What you get**:
+- Test case title and description
+- Priority level (Critical/High/Medium/Low)
+- Preconditions
+- Step-by-step test steps
+- Expected results
+- Appropriate tags (@smoke, @regression, @p0, etc.)
+
+**Example interaction**:
+```
+You: @test-case-generator
+     [Attach: login-feature-spec.pdf]
+     "Generate test cases for project PR-12345"
+
+Agent: I'll generate comprehensive test cases for your login feature.
+
+       Test cases generated:
+       ✓ TC-001: Login - Valid Credentials - Success (@smoke, @p0)
+       ✓ TC-002: Login - Invalid Password - Error (@regression, @p1)
+       ✓ TC-003: Login - Google OAuth - Success (@smoke, @p0)
+       ✓ TC-004: Login - Empty Fields - Validation (@regression, @p2)
+       
+       Added 4 test cases to Test Management project PR-12345
+```
+
+---
+
+## Real Examples
+
+### Example 1: Cross-Browser Bug Fix
+
+**Scenario**: Users report a broken layout on Safari.
+
+```
+💬 "Open localhost:3000/dashboard on Safari 17 on macOS Sonoma"
+   → BrowserStack opens a real Safari session
+   → You see the layout is indeed broken
+   
+💬 "Take a screenshot of this page"
+   → Screenshot saved to your project
+   
+💬 "Now open the same page on Chrome to compare"
+   → Opens Chrome session
+   → Layout works fine
+   
+💬 "What CSS properties might cause this Safari-specific issue?"
+   → Cursor analyzes your code and suggests the problem
+```
+
+### Example 2: Mobile App Crash Debug
+
+**Scenario**: Your iOS app crashes on launch for some users.
+
+```
+💬 "Open my app on iPhone 14 with iOS 16"
+   (Provide path: /Users/me/builds/app-v2.1.ipa)
+   → App launches successfully
+   
+💬 "Try on iPhone 15 with iOS 17"
+   → App crashes!
+   
+💬 "Get the crash logs for this session"
+   → Cursor retrieves logs
+   
+💬 "Analyze these crash logs and suggest a fix"
+   → Cursor identifies the issue and suggests code changes
+```
+
+### Example 3: Automated Test Failure Investigation
+
+**Scenario**: Your CI pipeline shows test failures on BrowserStack.
+
+```
+💬 "My Automate build 'feature-branch-123' has 3 failed tests. What went wrong?"
+   → Cursor fetches error logs and screenshots
+   
+💬 "Show me screenshots from the failed tests"
+   → Screenshots displayed
+   
+💬 "The login button selector seems wrong. Update my test file with the correct selector."
+   → Cursor updates your test file
+   
+💬 "Re-run the tests"
+   → Tests pass ✅
+```
+
+## What's Available
+
+**Manual Testing**
+- Test websites on any browser/OS combination
+- Test mobile apps on 3,500+ real devices
+- Access localhost from cloud devices (no deployment needed)
+
+**Automated Testing**  
+- Run Playwright, Selenium, Cypress, and more
+- Parallel test execution across browsers
+- Get screenshots, videos, and logs automatically
+
+**Test Management**
+- Create and organize test cases
+- Track test execution and results
+- Generate test reports
+
+**Accessibility Testing**
+- WCAG 2.0/2.1/2.2 compliance scanning
+- AI-powered fix suggestions
+- Local and production site scanning
+
+**AI-Powered Features**
+- Generate test cases from requirements documents
+- Auto-heal flaky test selectors
+- Convert manual tests to automation
+- Intelligent failure analysis and debugging
+
+## Troubleshooting
+
+**Plugin not responding?**
+- Verify your credentials are set correctly: `echo $BROWSERSTACK_USERNAME`
+- Restart Cursor after adding credentials
+- Check Node.js version: `node --version` (need 18+)
+
+**"Authentication failed" error?**
+- Double-check your username and access key from [Account Settings](https://www.browserstack.com/accounts/profile/details)
+- Make sure there are no extra spaces in your environment variables
+
+**Can't access localhost?**
+- This works automatically! Just use `localhost:PORT` in your prompts
+- BrowserStack creates a secure tunnel to your local machine
+
+**Behind a corporate firewall?**
+- Contact your IT team to allowlist BrowserStack domains
+- Or ask us about enterprise firewall configurations
+
+## Get Help
+
+- **Issues with the plugin**: [Open a GitHub issue](https://github.com/browserstack/mcp-server/issues)
+- **BrowserStack platform questions**: [Contact Support](https://www.browserstack.com/contact)
+- **Documentation**: [BrowserStack Docs](https://www.browserstack.com/docs)
+
+## Pricing
+
+- **Free Trial**: Test drive with limited minutes
+- **Open Source**: Free for qualifying projects ([apply here](https://www.browserstack.com/open-source))
+- **Paid Plans**: From $29/month for individuals, custom enterprise pricing
+
+See [pricing details](https://www.browserstack.com/pricing)
diff --git a/agents/test-case-generator.md b/agents/test-case-generator.md
new file mode 100644
index 0000000..2221ba0
--- /dev/null
+++ b/agents/test-case-generator.md
@@ -0,0 +1,95 @@
+---
+name: test-case-generator
+description: Generate comprehensive test cases from product requirements, user stories, or PRD documents using BrowserStack AI.
+---
+
+# Test Case Generator
+
+You help users generate test cases from requirements using BrowserStack's AI-powered test case generation.
+
+## When to Use
+
+- User provides product requirements or user stories
+- User uploads a PRD document
+- Need to create test cases for new features
+
+## Process
+
+### 1. Get Requirements
+
+Ask user to provide:
+- Text description of the feature
+- Or path to PRD file (PDF, image)
+
+### 2. Generate Test Cases
+
+Use BrowserStack AI to generate test cases:
+```
+"Generate test cases for [feature description]"
+"Generate test cases from PRD file at [file path]"
+```
+
+### 3. Organize in Test Management
+
+Create project structure:
+```
+"Create Test Management project called '[Project Name]'"
+"Add folder '[Feature Name]' to project"
+```
+
+### 4. Review and Add Test Cases
+
+The AI will generate test cases with:
+- Test case title
+- Steps to execute
+- Expected results
+- Priority level
+
+Review and add to Test Management:
+```
+"Add these test cases to Test Management project"
+```
+
+## Example Workflow
+
+**User**: "Generate test cases for a login feature with email and Google OAuth"
+
+**Your approach**:
+```
+I'll generate comprehensive test cases for your login feature.
+
+First, let me understand the requirements:
+- Email/password login
+- Google OAuth login
+- Forgot password flow
+- Any specific error scenarios?
+
+Then I'll use BrowserStack AI to generate test cases covering:
+- Happy path (successful login)
+- Error cases (wrong password, unregistered email)
+- OAuth flow
+- Edge cases (empty fields, SQL injection attempts)
+
+Once generated, I'll organize them in Test Management:
+"Create Test Management project 'User Authentication'"
+"Add test cases for email login scenarios"
+"Add test cases for OAuth login scenarios"
+```
+
+## Test Case Structure
+
+Each generated test case includes:
+- **Title**: Clear description (e.g., "Login - Valid Credentials - Success")
+- **Priority**: Critical/High/Medium/Low
+- **Preconditions**: What needs to be set up
+- **Steps**: Numbered actions to perform
+- **Expected Results**: What should happen
+- **Tags**: @smoke, @regression, @p0, etc.
+
+## Tips for Better Test Cases
+
+- Be specific about requirements
+- Include edge cases and error scenarios
+- Mention any integrations (third-party services)
+- Specify target platforms (web, mobile, specific browsers)
+- Note any compliance requirements (security, accessibility)
diff --git a/mcp.json b/mcp.json
new file mode 100644
index 0000000..d0568c8
--- /dev/null
+++ b/mcp.json
@@ -0,0 +1,12 @@
+{
+  "mcpServers": {
+    "browserstack": {
+      "command": "npx",
+      "args": ["-y", "@browserstack/mcp-server@latest"],
+      "env": {
+        "BROWSERSTACK_USERNAME": "${BROWSERSTACK_USERNAME}",
+        "BROWSERSTACK_ACCESS_KEY": "${BROWSERSTACK_ACCESS_KEY}"
+      }
+    }
+  }
+}
diff --git a/skills/accessibility-fix/SKILL.md b/skills/accessibility-fix/SKILL.md
new file mode 100644
index 0000000..cddcbdf
--- /dev/null
+++ b/skills/accessibility-fix/SKILL.md
@@ -0,0 +1,104 @@
+---
+name: accessibility-fix
+description: Scan a webpage for accessibility issues, identify violations, generate code fixes, and verify the fixes work. Use when fixing WCAG compliance or accessibility bugs.
+---
+
+# Accessibility Fix Workflow
+
+## When to Use
+
+- User wants to fix accessibility issues on a page
+- Need to ensure WCAG compliance
+- Fixing specific accessibility bugs
+
+## Steps
+
+### 1. Run Accessibility Scan
+
+Ask for the URL to scan:
+```
+"Run accessibility scan on [URL]"
+```
+
+Example: `"Run accessibility scan on localhost:3000"`
+
+### 2. Analyze Results
+
+Review the scan results and categorize issues:
+- **Critical**: Blocks users (missing form labels, keyboard traps)
+- **High**: Major barriers (poor color contrast, missing alt text)
+- **Medium**: Best practices (heading hierarchy)
+
+### 3. Fix Issues in Code
+
+For each issue, provide specific code fix:
+
+**Missing Alt Text:**
+```html
+<!-- Before -->
+<img src="logo.png">
+
+<!-- After -->
+<img src="logo.png" alt="Company Logo">
+```
+
+**Poor Color Contrast:**
+```css
+/* Before - Contrast 2.8:1 (fails) */
+color: #999;
+background: #fff;
+
+/* After - Contrast 4.6:1 (passes) */
+color: #666;
+background: #fff;
+```
+
+**Missing Form Label:**
+```html
+<!-- Before -->
+<input type="email" placeholder="Email">
+
+<!-- After -->
+<label for="email">Email Address</label>
+<input type="email" id="email" aria-required="true">
+```
+
+**Keyboard Navigation:**
+```html
+<!-- Before -->
+<div onclick="handleClick()">Click me</div>
+
+<!-- After -->
+<button onclick="handleClick()">Click me</button>
+```
+
+### 4. Verify Fixes
+
+After user implements fixes:
+```
+"Re-scan [URL] to verify accessibility issues are fixed"
+```
+
+### 5. Report Results
+
+Confirm what was fixed and any remaining issues.
+
+## Quick Reference
+
+**Common WCAG Requirements:**
+- Images need alt text
+- Color contrast ≥ 4.5:1 for text
+- All interactive elements keyboard accessible
+- Forms have labels
+- Proper heading hierarchy (h1 → h2 → h3)
+
+## Example
+
+**User**: "Fix accessibility issues on my login page at localhost:3000/login"
+
+**Workflow**:
+1. Scan: `"Run accessibility scan on localhost:3000/login"`
+2. Identify: "Found 3 critical issues: missing form labels, poor contrast, no keyboard nav"
+3. Fix: Provide code fixes for each issue
+4. Verify: `"Re-scan localhost:3000/login"`
+5. Confirm: "All critical issues resolved ✓"
diff --git a/skills/visual-issue-fix/SKILL.md b/skills/visual-issue-fix/SKILL.md
new file mode 100644
index 0000000..64989df
--- /dev/null
+++ b/skills/visual-issue-fix/SKILL.md
@@ -0,0 +1,130 @@
+---
+name: visual-issue-fix
+description: Capture screenshots of a page across browsers, identify visual issues, fix them in code, and validate. Use for cross-browser layout bugs or visual regressions.
+---
+
+# Visual Issue Fix Workflow
+
+## When to Use
+
+- Layout broken in specific browser
+- Visual bugs reported by users
+- Cross-browser design inconsistencies
+- Responsive design issues
+
+## Steps
+
+### 1. Capture Screenshots
+
+Open page on target browsers:
+```
+"Open [URL] on Chrome"
+"Open [URL] on Safari"
+"Open [URL] on Firefox"
+```
+
+Take screenshots:
+```
+"Take screenshot of the current page"
+```
+
+### 2. Identify Issues
+
+Compare screenshots and identify:
+- Layout differences (alignment, spacing)
+- Missing elements
+- Broken styling (fonts, colors, borders)
+- Responsive breakpoint problems
+
+### 3. Fix in Code
+
+Provide specific code fixes:
+
+**Flexbox Issue:**
+```css
+/* Before - Safari issue */
+.container {
+  display: flex;
+  gap: 20px;
+}
+
+/* After - Works everywhere */
+.container {
+  display: flex;
+}
+.container > * {
+  margin-right: 20px;
+}
+```
+
+**CSS Grid Issue:**
+```css
+/* Before - IE11 doesn't support */
+.grid {
+  display: grid;
+  grid-template-columns: repeat(3, 1fr);
+}
+
+/* After - Fallback for old browsers */
+.grid {
+  display: flex;
+  flex-wrap: wrap;
+}
+@supports (display: grid) {
+  .grid {
+    display: grid;
+    grid-template-columns: repeat(3, 1fr);
+  }
+}
+```
+
+**Responsive Issue:**
+```css
+/* Before - Fixed width breaks on mobile */
+.sidebar {
+  width: 300px;
+}
+
+/* After - Responsive */
+.sidebar {
+  width: 100%;
+}
+@media (min-width: 768px) {
+  .sidebar {
+    width: 300px;
+  }
+}
+```
+
+### 4. Validate Fix
+
+After user implements fix:
+```
+"Re-open [URL] on [browser] to verify the fix"
+"Take screenshot to confirm layout is correct"
+```
+
+### 5. Test Other Browsers
+
+Ensure fix didn't break other browsers:
+```
+"Test on Chrome, Firefox, Safari, and Edge"
+```
+
+## Quick Tips
+
+- Test on real devices, not just dev tools
+- Check both desktop and mobile
+- Verify at different screen sizes
+- Look for vendor prefix needs (-webkit-, -moz-)
+
+## Example
+
+**User**: "My navigation is broken in Safari"
+
+**Workflow**:
+1. Capture: `"Open localhost:3000 on Safari"` + `"Take screenshot"`
+2. Identify: "Navigation items overlapping - flexbox gap issue"
+3. Fix: Provide CSS fix replacing gap with margins
+4. Validate: `"Re-open localhost:3000 on Safari"` - confirm fixed
+5. Cross-check: Test on Chrome, Firefox to ensure no regression