Skip to main content
No Result Found
Get your setup working faster. Join our Discord for optimisation tips from elite testers. Join our DiscordJoin our Discord

Supported features and limits in Test Companion Cloud

Check the integrations, test management tools, frameworks, browsers, devices, and platform limits that Test Companion Cloud supports.

  • Availability: Test Companion Cloud is in beta and is not available to every account yet. It is available on request. To have it enabled for your account, contact BrowserStack support.
  • IDE: Test Companion Cloud works in Visual Studio Code only.

Use this page to check whether Test Companion Cloud fits your project before you start a task. When Test Companion Cloud does not support a feature, the last section on this page names the alternative.

At a glance

This table summarizes what Test Companion Cloud supports in each area:

Area Supported Not supported
IDE Visual Studio Code JetBrains IDEs, Visual Studio, Eclipse
Source control GitHub (github.com) GitLab, Bitbucket, Azure Repos, GitHub Enterprise Server
Requirements Jira Cloud, Jira Server and Data Center, Azure Boards, Confluence Cloud, Figma, GitHub pull requests Confluence Data Center, Azure DevOps Server, Linear, Slack
Test management BrowserStack Test Management, TestRail, Xray Cloud, Zephyr Scale Cloud, Azure Test Plans qTest, QMetry, Zephyr Squad, Xray Server and Data Center
Test reporting BrowserStack Test Reporting & Analytics GitHub Actions and Jenkins run URLs
Web browsers Chromium Firefox, Safari, WebKit, Edge
Mobile apps Appium on real Android and iOS devices Espresso, XCUITest, Maestro, Flutter integration tests, Detox
Accessibility Website and app accessibility scans None
Visual testing Percy and App Percy None

Requirements and project management tools

Paste a ticket, page, or design link in your prompt. Test Companion reads it as context to generate test cases, automate tests, and fix failures. Connect each tool on the Integrations page of your BrowserStack account first.

Tool Capabilities Requirements
Jira Cloud Read issues by link or key, including linked issues and Confluence pages. File bugs that Test Companion finds, after you approve each one. Jira connected to your BrowserStack account.
Jira Server and Data Center Read issues you paste in the prompt. Jira connected as self-hosted, and Visual Studio Code open on a network that can reach your Jira server.
Azure Boards Read work items by link. File bugs that Test Companion finds, after you approve each one. Azure DevOps connected. Organizations on dev.azure.com or *.visualstudio.com only.
Confluence Cloud Read pages by link. Confluence connected.
Figma Read designs and frames to generate test cases for websites. Figma connected.
GitHub pull requests Read the description and code changes of a pull request you link. GitHub connected.
Local changes Send your uncommitted diff as read-only context by typing @git-changes in the prompt. A Git repository open in Visual Studio Code.
Attachments Read images and documents you attach to a message. Up to 20 files and 100 MB per message.

Test management tools

Test Companion Cloud works with these test management tools:

Tool Read test cases Add or update test cases Generate test cases
BrowserStack Test Management Yes Yes Yes
TestRail Yes Yes Not yet
Xray Cloud Yes Yes Not yet
Zephyr Scale Cloud Yes Yes Not yet
Azure Test Plans Yes Yes Not yet

Test Companion Cloud automates test cases from these tools as follows:

Tool Automate selected test cases Update automation status
BrowserStack Test Management Yes Yes
TestRail From a prompt No
Xray Cloud Yes No
Zephyr Scale Cloud Yes No
Azure Test Plans Yes Yes

Test Reporting & Analytics

Test Companion Cloud fixes failed and flaky tests from BrowserStack Test Reporting & Analytics builds. You can start from the Failure Analysis panel. You can also give Test Companion a build URL, build ID, test run ID, or project and build name. Test Companion Cloud cannot read GitHub Actions or Jenkins run URLs, or import Allure and JUnit reports.

Web testing

Test Companion Cloud tests websites in Chromium, with a visible window at 1920 by 1080 pixels. Firefox, Safari, WebKit, and Edge are not available on the cloud machine. To run tests on other browsers, keep a browserstack.yml file in your repository. Test Companion then offers to run the tests on BrowserStack Automate.

Framework Status Notes
Playwright (JavaScript, TypeScript, Python, Java, .NET) Supported Test Companion Cloud installs the browser that matches your Playwright version.
Cypress Supported Runs in Chrome or Electron.
Selenium (Java, Python, JavaScript) Supported Uses the Chrome browser on the cloud machine.
WebdriverIO, Puppeteer Supported Uses the Chrome browser on the cloud machine.
TestNG, JUnit, pytest, Jest, Vitest Supported Runs through the build tool of your project.
Selenium with C#, Robot Framework Not verified The tools are available on the cloud machine, but these frameworks are not tested yet.

Test Companion Cloud sets up the runtime for the language your project uses. Supported languages are Node.js, Python, Java, Go, Ruby, .NET, Rust, PHP, and Elixir. It reads the version from files such as .nvmrc, .node-version, .python-version, .java-version, .ruby-version, .tool-versions, pom.xml, build.gradle, go.mod, or global.json. During setup, the cloud machine installs JavaScript dependencies. Test Companion installs the other dependencies during the task.

Mobile app testing

Test Companion Cloud tests native apps on real Android and iOS devices on BrowserStack. App tasks run under your App Automate plan.

Capability Status Notes
Native apps on real Android and iOS devices Supported Upload an .apk, .aab, .xapk, or .ipa file. Requires an App Automate plan.
Appium scripts in Java, JavaScript (WebdriverIO), and Python Supported N/A
Live device stream and session recordings Supported Shown in the Device tab of the task.
Private or internal apps Supported Turn on local testing.
Device settings Supported Orientation, permissions, location, media injection, and one-time passwords.
Several devices in parallel Supported Up to the number of parallel sessions in your plan, and at most eight subtasks per task.
React Native and Flutter apps, tested as built apps Supported Test Companion tests the installed app with Appium.
Espresso, XCUITest, Maestro, Flutter integration tests, Detox Not supported Use local mode.
Mobile websites on real devices Not supported Use local mode.
Several devices in one subtask Not supported Each subtask uses one device.

Accessibility and visual testing

Test Companion Cloud can run these accessibility and visual checks:

Capability Status Notes
Website accessibility scan with base rules Supported Public URLs, and localhost with local testing on.
Website accessibility scan with Advanced + AI rules Supported One combined report for several pages, on your BrowserStack Accessibility dashboard. Works on localhost with local testing on. Pages that need a login are not scanned.
App accessibility scan Supported On real devices.
WCAG versions and levels Supported WCAG 2.1 and 2.2, at level A, AA, or AAA. The default is WCAG 2.1 AA.
Fix accessibility issues Supported When you ask. The fixes arrive as a pull request.
Percy visual tests for websites Supported Runs under your BrowserStack account. The Visual Analysis panel opens on builds that Test Companion starts during the task, not on builds that a subtask starts.
App Percy visual tests for native apps Supported N/A

Source control

Test Companion Cloud works with repositories as follows:

Capability Status
Public and private repositories on github.com Supported
Start from any branch you choose in the workspace Supported
Install only the test folders you choose, in a monorepo Supported
Pull request opened when the task finishes, and updated by follow-ups Supported
Tasks without a repository Supported
Unpushed commits and uncommitted changes Not supported. Push your changes first, or type @git-changes in the prompt to send your uncommitted diff as read-only context.
GitLab, Bitbucket, Azure Repos, GitHub Enterprise Server Not supported

Customization

In Cloud mode, Test Companion uses the rules, skills, and agents you turned on, up to 5 MB in total. It also remembers what it learns about a repository from one task to the next. It keeps this memory separately for each user and repository.

These are not available in Cloud mode yet: MCP servers, model selection, and Coverage Analyzer. Multi-user browser testing with more than one browser profile is also not available.

Platform limits

Every task in Cloud mode runs within these limits:

Item Limit
Machine in the BrowserStack cloud Linux, 4 vCPUs, 8 GB memory, 50 GB disk
Subtasks for each task 8. When fewer parallel slots are free in your organization, Test Companion Cloud runs the subtasks it can start.
Tasks running at the same time, for each organization 20, including the subtasks within each task
Task with no activity Ends after 90 minutes. There is no limit on the total length of a task.
Question with no answer The task ends after 20 minutes. Send a message to continue it.
Plan with no approval The task waits up to 90 minutes.
Attachments 20 files and 100 MB per message, 20 MB per file
Workspaces 50 for each user
Usage Tasks in Cloud mode use the same AI credits as local mode.

Not supported in Cloud mode

These features are not available in Cloud mode. Where one exists, the table names the alternative:

Feature Alternative
JetBrains IDEs, Visual Studio, Eclipse Use Cloud mode in Visual Studio Code, or run the task in local mode in your IDE.
Firefox, Safari, WebKit, Edge Run the task in local mode, or run your tests on BrowserStack Automate with browserstack.yml.
GitLab, Bitbucket, Azure Repos, GitHub Enterprise Server Run the task in local mode.
Espresso, XCUITest, Maestro, Flutter integration tests, Detox Run the task in local mode.
Mobile websites on real devices Run the task in local mode.
qTest, QMetry, Zephyr Squad, Xray Server and Data Center No alternative in Cloud mode.
Confluence Data Center, Azure DevOps Server Attach the content to your message as a file.
GitHub Actions or Jenkins run URLs as a failure source Report the results to Test Reporting & Analytics and start from the Failure Analysis panel.
MCP servers Run the task in local mode.
Allure and JUnit report import Run the task in local mode.
Linear, Slack Paste the content in your prompt.

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

Is this page helping you?

Yes
No

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!

Talk to an Expert
Download Copy Check Circle