Skip to main content
Agents are the fundamental building blocks of Iris’s testing capabilities. They are powered by Large Language Models (LLMs) and equipped with specialized tools that enable them to interact with web applications just like human users would.

​
What are Iris Agents?

Language Understanding

Process and understand complex instructions and website content

Visual Processing

Analyze and interact with UI elements and layouts

Navigation

Autonomously explore and interact with web applications

Testing

Execute and validate test scenarios intelligently

​
Key Features

​
Visual & Interactive Capabilities

Process and understand website layouts and UI elements
Comprehend textual content and error messages
Perform complex user interactions naturally

​
Creating an Agent

1

Define Purpose

Specify the agent’s testing objectives and scope
2

Configure Permissions

Set up appropriate access levels and capabilities
3

Testing Parameters

Configure test-specific settings and behaviors
4

Autonomous Settings

Define autonomous mode preferences if required

​
Best Practices

  • Start with clearly defined test objectives
  • Plan agent interactions carefully
  • Document expected behaviors
  • Configure appropriate permission levels
  • Set up meaningful test parameters
  • Define clear success criteria
  • Monitor agent activities during runs
  • Track performance metrics
  • Review and analyze results

​
Use Cases

Automated Testing

End-to-end testing, user journey validation, and regression testing

UX Analysis

Navigation assessment, UI/UX evaluation, and performance monitoring

Documentation

Content verification, link validation, and cross-referencing