Ask the Community
Test lifecycle
This section covers how to use Test Companion to create, view, and organize your test cases directly within your IDE.
Test Companion uses a set of specialized AI Agents to perform these tasks. You select an agent from the chat dropdown based on the job you want to do. Once your extension is installed, and you are signed in, you can start by generating new test cases with AI or by managing your existing test repository.
What are AI agents?
Agents are AI-powered specialists, each trained for a specific task. You will use different agents to:
- Generate new test cases from requirements
- Explore a website to find test scenarios
- Convert manual tests into automated scripts
- Analyze and fix failing tests
You can select an agent from the chat dropdown in the Test Companion panel. They are grouped into two categories:
Core Agents
Core agents are general-purpose assistants that can handle a broad range of tasks based on your end-to-end testing goals or immediate developer context.
| Agent | What it does |
|---|---|
| Tester | Your primary agent for the complete testing lifecycle. Use this for broad, end-to-end tasks like generating a full test plan from a large document. |
| Dev Tester | A developer-focused agent that understands your local code. Use this to test your current changes, analyze a new feature, or write tests in your immediate context. |
Specialized Agents
Specialized agents are trained to perform one specific task with high accuracy. Use them when you know exactly what you want to do, as they provide more focused results.
| Agent | What it does |
|---|---|
| Test Case Generator | Reads requirement documents (like a PRD or spec) to write and modify test cases. |
| Exploratory Tester | Analyzes a live website or URL to discover potential test scenarios and write test cases for them. |
| UI Test Generator | Converts existing manual test cases into executable UI automation scripts (e.g., Playwright, Cypress). |
| API Test Generator | Generates API tests for a specific endpoint, URL, or API specification. |
| Test Fixer | Analyzes test failures from your last build and provides suggested code fixes to debug them. |
Understand the AI settings
Next to the prompt bar, you will see several settings that control the AI’s behavior.
- Auto-approve: When Enabled, the AI will automatically add the generated test cases to your list. If disabled, you will be prompted to review and approve them first.
- Enabled / Disabled: This toggle turns the AI generation feature on or off.
- Read: This feature is for reading existing files for context (e.g., Read: settings.json).
- Edit: Click this to modify the context files the AI is reading.
When you click the expand icon, you will see a full list of permissions you can grant the AI. This section controls what the AI is allowed to do on its own when Auto-approve is enabled.
| Setting | What it does |
|---|---|
| Enable auto-approve | The main toggle. This must be checked for any of the other actions to work automatically. |
| Toggle all | A helper to quickly check or uncheck all actions in the list. |
| Read project files | Allows the AI to automatically read files inside your current project workspace. |
| Read all files | Allows the AI to read any file on your computer, even those outside your project. |
| Edit project files | Allows the AI to automatically edit or modify files inside your current project. |
| Edit all files | Allows the AI to edit any file on your computer. |
| Execute safe commands | Allows the AI to run safe, read-only commands in your terminal (like ls or git status). |
How-to guides
This section walks you through the complete testing lifecycle.
- Generate test cases with AI: Learn how to create new test suites from a PRD, spec document, or a URL.
- Manage test cases: Explore the UI for organizing, editing, and syncing your tests with Test Management.
- Automate tests with AI: Convert existing manual test cases into executable automation scripts.
- Fix failed tests: Use AI to analyze test failures from your last build and get suggested fixes.
We're sorry to hear that. Please share your feedback so we can do better
Contact our Support team for immediate help while we work on improving our docs.
We're continuously improving our docs. We'd love to know what you liked
We're sorry to hear that. Please share your feedback so we can do better
Contact our Support team for immediate help while we work on improving our docs.
We're continuously improving our docs. We'd love to know what you liked
Thank you for your valuable feedback!