Test Execution Guide¶

Learn how to leverage TINAA's AI-powered test execution and browser automation capabilities.

Overview¶

TINAA provides intelligent browser automation through MCP (Model Context Protocol) integration, enabling AI assistants like Claude to execute Playwright tests and perform browser automation tasks.

Test Execution Methods¶

Via HTTP API¶

Execute predefined test types through the HTTP API:

# Exploratory Testing
curl -X POST http://localhost:8765/test/exploratory \
  -H "Content-Type: application/json" \
  -d '{
    "action": "exploratory",
    "parameters": {
      "url": "https://example.com",
      "focus_area": "navigation"
    }
  }'

# Accessibility Testing
curl -X POST http://localhost:8765/test/accessibility \
  -H "Content-Type: application/json" \
  -d '{
    "action": "accessibility",
    "parameters": {
      "url": "https://example.com",
      "standard": "WCAG2AA"
    }
  }'

Via MCP Integration¶

When integrated with Claude Desktop, TINAA provides these tools:

navigate_to_url - Navigate browser to any URL
take_page_screenshot - Capture screenshots
click_element - Interact with page elements
fill_form_field - Fill form inputs
run_exploratory_test - Execute AI-guided exploratory tests
check_page_accessibility - Run accessibility audits

Available Test Types¶

1. Exploratory Testing¶

TINAA uses AI heuristics to explore your application:

{
  "action": "exploratory",
  "parameters": {
    "url": "https://example.com",
    "depth": 3,
    "focus_area": "forms"
  }
}

2. Accessibility Testing¶

Validate WCAG compliance:

{
  "action": "accessibility", 
  "parameters": {
    "url": "https://example.com",
    "standard": "WCAG2AA",
    "includeWarnings": true
  }
}

3. Responsive Testing¶

Test across multiple viewports:

{
  "action": "responsive",
  "parameters": {
    "url": "https://example.com",
    "viewports": ["mobile", "tablet", "desktop"]
  }
}

4. Security Testing¶

Basic security checks:

{
  "action": "security",
  "parameters": {
    "url": "https://example.com",
    "checks": ["https", "headers", "forms"]
  }
}

Test Patterns and Resources¶

TINAA includes predefined test patterns located in /resources/:

Test Patterns (test_patterns.json) - Common testing scenarios
Accessibility Rules (accessibility_rules.json) - WCAG compliance rules
Security Patterns (security_test_patterns.json) - Security check definitions
Exploratory Heuristics (exploratory_heuristics.json) - AI exploration rules

Writing Custom Test Scripts¶

TINAA can analyze and execute custom Playwright scripts:

// my-test.js
const { test, expect } = require('@playwright/test');

test('my custom test', async ({ page }) => {
  await page.goto('https://example.com');
  await expect(page).toHaveTitle(/Example/);
});

Analyze your script:

tina analyze my-test.js

Integration with AI Assistants¶

When used with Claude Desktop, you can:

Ask Claude to navigate to websites
Request screenshots of specific pages
Have Claude perform exploratory testing
Get accessibility reports
Execute custom test scenarios

Example prompts: - "Navigate to example.com and take a screenshot" - "Check the accessibility of this login form" - "Perform exploratory testing on the checkout flow"

Real-time Progress Tracking¶

TINAA provides WebSocket connections for real-time test progress:

const ws = new WebSocket('ws://localhost:8765/ws');

ws.on('message', (data) => {
  const progress = JSON.parse(data);
  console.log(`${progress.action}: ${progress.status}`);
});