LocatAI 1.3.1-beta

LocatAI

AI-powered element locator for Selenium WebDriver automation. Supports multiple AI providers including OpenAI, Anthropic, Google Gemini, and Ollama.

Installation

Install the package from NuGet:

dotnet add package LocatAI

Configuration

1. Create an appsettings.json file in your project with the following structure:

{
  "AI": {
    "Provider": "openai",
    "ApiKey": "your-api-key-here",
    "Model": "gpt-4"
  },
  "Options": {
    "MinimumConfidence": 0.65,
    "MaxLocatorsToTry": 5,
    "TimeoutMilliseconds": 10000,
    "EnableDetailedLogging": true,
    "EnableLocatorRetry": true,
    "MaxRetryAttempts": 2
  }
}

Set the "Provider" to one of:

• openai (requires OpenAI API key)
• anthropic (requires Anthropic API key)
• gemini (requires Google AI Studio API key)
• ollama (requires Ollama running locally, set ApiKey to Ollama URL e.g. "http://localhost:11434")

2. Initialize LocatAI in your code:

using Microsoft.Extensions.Configuration;
using LocatAI.Core;

// Load configuration from appsettings.json
var config = new ConfigurationBuilder()
    .SetBasePath(AppContext.BaseDirectory)
    .AddJsonFile("appsettings.json", optional: false, reloadOnChange: true)
    .Build();

// Initialize ElementFinder with configuration
ElementFinder.Initialize(config);

3. Use the extension methods to find elements:

// Basic usage - returns just the IWebElement
IWebElement username = driver.FindElementByLocatAI("username input field");
IWebElement loginButton = driver.FindElementByLocatAI("login button");

// Advanced usage - returns both element and locator details
var (element, locator) = driver.FindElementWithDetailsByLocatAI("submit button");
Console.WriteLine($"Found element using {locator.LocatorType}={locator.LocatorValue} with confidence {locator.Confidence:P}");

// Async versions available
var searchBox = await driver.FindElementByLocatAIAsync("search input box");
var (submitBtn, locatorInfo) = await driver.FindElementWithDetailsByLocatAIAsync("submit or search button");

Features

• Natural language element location using AI
• Multiple AI provider support (OpenAI, Anthropic, Google Gemini, Ollama)
• Smart caching of successful locators
• Automatic retry with fallback strategies
• Detailed logging options
• Configurable confidence thresholds and timeouts
• Both synchronous and asynchronous methods
• Supports all major locator types:
  • ID
  • Name
  • CSS Selector
  • XPath
  • Class Name
  • Tag Name
  • Link Text
  • Partial Link Text

Advanced Features

Confidence Scores

Each locator comes with a confidence score (0-1) indicating how likely it is to work. Configure minimum confidence with MinimumConfidence option.

Smart Caching

Successful locators are cached based on URL, element description, and DOM hash. This improves performance on subsequent visits.

Automatic Retry

If enabled with EnableLocatorRetry, failed locators trigger new AI-generated alternatives up to MaxRetryAttempts times.

Smart Waits

The system learns optimal wait times based on historical timing data for each element, automatically adjusting implicit waits.

Performance Tracking

Failed locators with low success rates are automatically removed from cache to maintain high reliability.

Dependencies

• Selenium.WebDriver (4.31.0+)
• Microsoft.Extensions.Configuration
• System.Text.Json
• One of the following AI providers:
  • OpenAI API key
  • Anthropic API key
  • Google AI Studio API key
  • Local Ollama installation

License

MIT