# Create test cases

> Create a test case, build its steps with the six step types, and manage the test case library.

A test case is a list of steps that tests one flow in your application. Write the steps as NLPs, record them with the Recorder extension or Copilot, or generate them with Atto.

## By application type

How you create a test case depends on the application type you test. This page covers what is common to all of them. Each application type has its own page for creating a test case:

| Application type | Where to start |
| --- | --- |
| Web and mobile web | [Create a web or mobile web test case](https://testsigma.com/docs/v2/application-types/web/create-web-test-case/) |
| Android | [Create an Android test case](https://testsigma.com/docs/v2/application-types/mobile/create-android-test-case/) |
| iOS | [Create an iOS test case](https://testsigma.com/docs/v2/application-types/mobile/create-ios-test-case/) |
| Unified mobile (Android and iOS) | [Create a Unified mobile test case](https://testsigma.com/docs/v2/application-types/mobile/unified/) |
| Windows desktop | [Create a Windows test case](https://testsigma.com/docs/v2/application-types/windows/create-windows-test-case/) |
| Salesforce | [Create a Salesforce test case](https://testsigma.com/docs/v2/application-types/salesforce/create-salesforce-test-case/) |
| REST and SOAP APIs | [Create an API test case](https://testsigma.com/docs/v2/application-types/rest-apis/create-api-test-case/) |

## Folders and subfolders

Test cases live in folders and subfolders. A folder groups related subfolders, and each subfolder holds its test cases. In a shopping application, **User Authentication** is a folder, **Login** is a subfolder, and **Login with invalid credentials** is a test case.

This structure keeps a large library searchable, prevents duplicate test cases, and feeds the filters you use when you build test suites.

To create a folder, click **+**, select **New Folder**, name it, and click **Add**. To create a subfolder, click **+**, select **New Subfolder**, select the parent folder in **Select Folder**, click **Next**, name the subfolder, and click **Create**.

![Test Case Explorer, with subfolders under a folder and the New Test Case control](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/Feature_Test_case_explorer.png)

For quick actions, click the ellipsis icon `⋮` on a folder or subfolder row:

| Action | What it does |
| --- | --- |
| Rename | Changes the name |
| Clone | Creates a copy |
| Move | Places it under a different folder |
| Delete | Removes it from the project |

Click **<** to collapse the folder list and give the test case more width.

---

## Create test steps

1. Click **+ Add new step**.

2. Enter the action in plain English. Suggestions appear as you type an action word.

3. Click **Create Step**.

![A test case's steps written as NLPs, with Add new step at the end](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_npl_2.png)

These action words have suggestions: `Click`, `Tap`, `Verify`, `Scroll`, `Check`, `Clear`, `Close`, `Uncheck`, `Store`, `Double Click`, `Drag`, `Enter`, `Swipe`, `Switch`, `Execute`, `Select`, `Wait`, and `Mouseover`.

Recording needs the [Recorder extension](https://chromewebstore.google.com/detail/testsigma-recorder/epmomlhdjfgdobefcpocockpjihaabdp) installed in your browser.

**Web and mobile web**

1. Create a first step carrying the URL to automate.

2. Click **Create Step**.

3. Click **Record**. A new window opens the URL.

4. Wait for the page to load fully, so the extension can collect the page information.

5. Perform the flow. Each action lands as a step in the recorder.

6. Click **Stop**. The test case page reopens with the recorded steps.

![The recorder panel over the application, with Generate Steps, Pause, and Stop](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/web_test_case_record_2.png)

**Android and iOS**

1. Click **Create Step**.

2. Click **Record**.

3. In the **Record test steps** overlay, select the Test Lab and the Test Machine.

4. Click **Upload** and supply the application build.

5. Click **Record**.

6. Perform the flow on the streamed device.

7. Stop the recording.

![The Record test steps overlay, with Test Lab, Test Machine, and App Source](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/Entities_for_Selection_in_Overlay.png)

If direct recording misses an action, use **Tap** to record the element instead.

The device panel and the element actions are the same on Android and iOS.

- **Device controls**: Install App, Mirroring mode, Inspect Mode (view element attributes without recording a step), Swipe By Coordinates, Tap By Coordinates, Search Element, Go back, Home, Hide Keyboard, Rotate Screen
- **Element actions**: Tap, Enter Data, View Code, Clear, Element Details

**Inside a WebView**

A blank screen with no selectable elements usually means a hybrid application rendering a WebView. Switch context to reach the elements inside it.

1. Click **H** in the recorder panel.

2. Select the WebView.

3. Perform the flow inside the WebView.

4. Switch back to the native context when you leave the WebView.

Each switch is recorded as a step:

- *Switch to Webview context* when you enter the WebView
- *Switch to Native App Context* when you return to the native application
- *Switch to context with name WEBVIEW_6890.1* when you pick a named view

Deleting a context switch step breaks playback inside the WebView. The behavior is the same on Android and iOS.

Copilot needs Testsigma Terminal running on your desktop. If it isn't running, the editor shows a prompt with **Launch Terminal**.

1. Enter the URL to automate, and click **Create Step**.

2. Click **Copilot**.

3. Click **Launch** in the **Copilot** overlay. Copilot opens a browser window and runs the URL.

4. Click **Rec**.

5. Perform the flow. Each action lands as a step in the background.

6. Click **Stop recording**.

7. Click **Exit Copilot**.

8. Click **Stop Copilot** in the **Stop & Exit Copilot** dialog.

Click **Generate Steps** in the Copilot dialog to have Copilot suggest test scenarios for the current page. Select a scenario, and Copilot writes its steps.

Atto's Generator agent turns requirements or an API schema into complete test cases, not single steps. Generated test cases are saved under the **AI Generated** folder and subfolder.

Generative AI features must be turned on under **Settings > Preferences**.

1. Go to Atto's Home and click **Generate with AI**.

2. Add your requirements as sources.

3. Describe what to cover in the prompt box.

4. Click **Generate with AI**.

5. Expand a category and open a generated test case.

6. On the **Manual Steps** tab, click **Edit** to change the steps directly, or enter a prompt and click **Refine manual steps**.

7. Click **Generate Automated Steps** to convert the manual steps to NLP steps.

8. (Optional) To check the test case before you save it, hover over **Run with Copilot** and select an environment.

9. Click **Save to Library**.

10. Select the folder and subfolder.

**Read existing test case library** is selectedby default, so Atto creates around the test cases you already have. Once it is selected in a session, it cannot be turned off again.

When the test case already exists, Atto updates it and marks it with an **Update** tag. Click **See What's New** to compare the versions. Saving then opens **Overwrite Test Case**, where you select **Overwrite**, **Save as New**, or **Link to original test case**.

**From an API schema**

Turn on **Generate test cases from Swagger schema** under **Settings > Preferences > Generative AI features**.

Atto groups REST endpoints into test cases by their Swagger tags. Tag every endpoint before you import.

1. In Test Case Explorer, click **Atto**.

2. Select **Generate Test Cases from API Schema**.

3. Import the schema. REST schemas use `.json`. SOAP schemas use `.wsdl` or `.xml`.

4. Clear the test cases you want to skip. All generated test cases are selected.

5. Review the steps in **Test Steps**, and the endpoint, body, and status verifications in **Verification Details**.

6. Click **Save Test Cases**.

Steps default to natural language. For loops, conditions, blocks, step groups, and API calls, see [Step types](#step-types).

## Manage test cases

Find test cases with filters and labels, delete and restore them, and import test cases from another project or from Postman.

### Find and organize test cases

Click **List View** to search, sort, and filter the whole library. Search by name from the search bar. Sort by **Title**, **Created Date**, or **Updated Date**. To filter, click **Show Filters > Add Filter**. Filters stack, and you can filter by Status, Priority, Last run result, Added to Test Suite, Assignee, Created By, Created Date, Updated Date, Reviewer, Test Case Type, Labels, Requirement, Requirement Type, and any custom field.

![List View with Add Filter open, showing the filter fields](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_list_actions_5.png)

To keep a filter, click **Save Filter > As New**, name it, and select **Mark as Public** to share it. **Replace Existing** updates a filter you already saved. Saved filters live under **Saved Filters**, split into Custom and Predefined, and you can edit or delete each one. Click **Reset** or **All** to clear the active filter.

### Label test cases

One label can be attached to test cases, step groups, elements, test suites, and test plans. To create a label, go to **Settings > Labels** and click **Add New Label**. To attach it to a test case, open the test case and use **Manage Test Case > Labels**. Step groups, elements, test suites, and test plans attach labels from their own panels.

To detach a label, go to **Settings > Labels**, click the label's count, select the assets in **Linked Entities**, and click **Remove Link**.

### Delete and restore test cases

Delete a test case 3 ways:

- Open the test case, click the **More options** `⋮` menu, and click **Delete**.
- Click the ellipsis icon `⋮` next to the test case in its subfolder, and click **Delete**.
- Select one or more test cases in List View, and click the **Delete** icon in the menu bar.

In **Delete Confirmation**, click **Delete** to remove the test case from the project.

Deleting a test case removes its associations with test suites, test plans, and prerequisites.

A deleted test case goes to the trash. To restore it, click **Saved Filters** in List View, select **Trash (Deleted Test Cases)**, find the test case, click **Restore** next to it, then click **Restore** in the dialog.

![The Trash filter in List View, with Restore and the permanent-delete icon on each row](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/Restore_Delete.png)

Deleting a test case permanently also removes its run reports and associated configurations. This cannot be undone.

To delete a test case permanently, click **Delete** next to it in the trash, enter `DELETE`, and click **I Understand, delete this (test-case-name)**.

### Import test cases from another project

Test cases import into the same application type only. Test cases from a web application import into a web application in the target project, and cannot move to a different application type.

1. Go to **Create Tests > Test Cases**.

2. In Test Case Explorer, click the ellipsis icon `⋮` and click **Import**.

3. On the **Select Source** tab of **Import Artefacts**, select the source **Project**, **Application Name**, and **Version**.

4. Click **Next**.

5. On the **Select Artefacts** tab, select the test cases, step groups, elements, test data profiles, environment variables, and uploads to import.

6. Click **Proceed to Import**.

7. On the **Start Import** tab, review the artefacts and click **Confirm**.

8. In **Create a Save Point?**, enter a name and click **Start Importing**.

9. Click **Done**.

![The Select artefacts step of Import Artefacts, with the artefact tabs and their counts](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_import_5.png)

Selecting a test case also selects its step groups, elements, test data profiles, environment variables, and uploads. To include artefacts that are not linked to the test cases you selected, open that artefact's tab and turn on **Show unlinked**. Turn on **Show selected** to review your selections on each tab.

On the **Start Import** tab, each test case carries a status: **New** if it does not exist in the target project, **No Change** if it exists and matches, or **Override** if it exists and has been modified.

Interrupting an import can cause data loss. Wait for it to finish. Import a smaller amount of data at a time, because a large import takes longer.

Testsigma creates a save point before the import, so you can revert the project to its state before the import. See Save points.

To check an import afterwards, go to **Settings > Imports**, click the import's status, and review the details in **Import Summary**.

### Import a Postman collection

1. Go to **Settings > Imports > Import**.

2. Select the exported JSON or ZIP file.

3. Select the project, application, and version.

4. Review the mapping preview.

5. Click **Start Importing**.

An import that matches an existing collection name creates a new entry with a timestamp appended.

Testsigma emails you when the import finishes. You can download the imported files to check them.

| Postman | Testsigma |
|---|---|
| Collection | Test suite |
| Subfolder | Test case |
| API inside a subfolder | Test steps |
| Folder inside a subfolder | A block inside the test case |
| Collection variables | Test data profiles |
| Global and environment variables | Environments |
| Parent folder | Test case label |

Test scripts, prerequisite scripts, settings, unsupported authorizations, unsupported HTTP methods, and GraphQL requests are not imported.

## Step types

A step is a natural language action by default. Change its type to reuse a sequence, repeat steps over test data, branch on a condition, group steps under a label, or call an API.

Click the option on the left side of a step to open the step type panel. Six types are available.

![The step type panel open on a step, listing the available types](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_step_type_3.png)

Loops, conditions, and blocks hold child steps, added with the **Step Inside** control. Child steps are numbered under their parent, such as **2.1** and **2.2**.

| Step type | What it does | How to add it | Limits and behavior |
|---|---|---|---|
| Natural language | Runs one action or assertion, written in plain English | Enter the action in the step | The default type |
| Step group | Calls a reusable sequence of steps from the test case | Select **Step Group**, select the group, then click **Create Step**. To build a group from steps you already have, select the steps and click **Create Step Group** | Groups nest up to 3 levels. **Create** copies the selected steps into a new group. **Create and Replace** also swaps them for the group call, and is offered only for consecutive steps. **Reuse Step Group** pulls a group from another project, application, or version; check its element locators against the target application first. Inside a test case you can edit a group step's test data and elements, but not its NLP. Click **Update Step** to save. These edits stay local to that test case |
| For loop | Repeats its child steps once per data set of a test data profile | Select the NLP variant for the data sets you want: the whole profile from start to end, an index range, sets filtered by name (contains, starts with, ends with, or IN), or sets filtered by a parameter value | Ends at the last data set, or at a break |
| While loop | Repeats its child steps while a condition holds, then moves on | Select the type, then add child steps with **Step Inside Loop** | Works in the recorder on web, mobile web, Android, and iOS |
| If condition | Runs its child steps only when the condition is true | Select **Conditional Step Types > If Natural Language**, then add child steps with **Step Inside IF**. Hover over the **IF** step to add **Else If** and **Else** | Also works while recording |
| Block | Labels consecutive steps as a named group, without making them reusable | Select **Create a Block**, or select consecutive steps and click **Create Block**. Click **>** to expand it, then add steps with **Step Inside Block** or **Step After Block** | A block can be reordered or moved within the test case. Deleting a block keeps its steps. Blocks cannot nest, cannot contain another block step at creation, and cannot be converted back to a plain step |

Inside a for loop, a while loop, or a step group, the `Store` NLP saves the current iteration count into a variable. The String Compare addon compares iterated values inside an if condition.

The panel also offers the REST API step type. See [REST API steps](#rest-api-steps).

### Run database queries from a step

The mysql_queries addon runs MySQL from NLP steps over a JDBC URL. Install it from the addons page.

The connection string takes this form:

```text
jdbc:mysql://<hostname>:<port>/<database>?user=<user>&password=<password>
```

Its NLPs cover:

- Executing a query, a create-procedure statement, a stored procedure call, an update, or a `.sql` script file
- Storing a select result into a variable
- Verifying a select result, or the affected-row count of a query
- Comparing two queries on one connection, or the same query across two connections

## REST API steps

A REST API step calls an API inside any test case. Configure the request, verify the response, and store values for the steps that follow. For API-only testing, create the project with the **Rest API** application type.

### Configure the request

1. Select **Rest API** from the step type panel.

2. Enter a **Title** for the step.

3. Enter the endpoint URL.

4. Select the method: **GET** to retrieve, **POST** to add, **PUT** to replace, **PATCH** to update fields, or **DELETE** to remove. The default is **GET**.

5. Fill in what the API needs:
   - **Parameters**: the URL field and the **Parameters** tab stay in sync. Enter `?name=Joel&type=new` in the URL, or enter key-value pairs in the tab. Path parameters use placeholders such as `/customer/:id`.
   - **Headers**: key-value pairs, such as `Accept: application/json`.
   - **Authorization**: **No Auth** by default. Select a type and fill in its fields.
   - **Body**: see the body types below.
   - **Settings**: request-specific options.

6. (Optional) Click **Send** to try the request live.

7. Click **Create**.

![A REST API step, with the method, endpoint, request tabs, and the response pane](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_step_type_RESTAPI_4.png)

**Global Objects** links stored objects and other predefined project objects into the request for reuse.

SOAP APIs use the same step. Send the XML envelope as the raw body with a `content-type: text/xml; charset=utf-8` header, using the **POST** method.

### Body types

- **None**: requests without a body. This is the default
- **form-data**: key-value pairs with a content type. A key holds text or a file
- **x-www-form-url-encoded**: key-value pairs encoded like URL parameters
- **Raw**: JSON, text, or XML, with syntax highlighting and automatic headers
- **Binary**: a single non-text file, such as an image or a video
- **GraphQL**: a query and optional variables, in JSON or table form

Form-data file paths persist across repeated calls. Uploading multiple files that each carry their own content type is not supported.

Selecting **GraphQL** sets the method to **POST**. A body on a **GET** request has no defined semantics, and some servers reject it.

Files sent as form-data or Binary also appear on the **Attachments** tab after you click **Create**.

### Verify the response

Verifications live on the response **Body**, **Headers**, and **Status**. Add them any of 4 ways:

- Send the request, click **Outline** on the response, and select **Add verification**
- Hover over an HTML response line and capture an attribute
- Add fields on the **Verification** tab, with a JSON or XPath path, an expected value, and a verification type. The **Status** tab takes a key name instead of a path
- Click **Copy Response**, paste the copied path into the path field, then set the type and the expected value

![The response Outline, with Store Variable and Add Verification on a field](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_verification_n_1.png)

**Verify Response Body** compares the whole body instead. Set **Comparison Type** and **Verification Type**. Before the API is invoked, it also asks for **Response Body Type** and an expected value.

Each verification type passes under a different condition:

- **Strict**: every condition matches exactly as specified
- **Strict Order**: the conditions match in the specified order
- **Lenient**: the essential conditions match, and the rest may be relaxed
- **Non-extensible**: only the pre-defined rules apply, with no extension
- **Schema**: the response satisfies a structural schema

### Store values for later steps

Stored variables capture parts of the response for use later in the test case or the session.

- **Body fields**: click **Outline > Store Variable**, or add them on the **Stored Variables** tab
- **HTML attributes**: hover over the response line and select the attribute
- **Headers**: click **Store Variable** on the **Headers** tab, or hover over a response header

**Save Response > As an Object** stores the whole response as a stored object instead. Stored objects have global scope, work across test cases, and can be downloaded.

### Use test data in a request

Any request field takes test data instead of a literal value.

1. Select the value to parameterize.

2. Click **Insert Test Data**.

3. Select a Parameter, Runtime, Environment, Random, Data Generator, Phone Number, or Mail Box value.

4. (Optional) Enter trial values under **Add Request Values**, click **Apply**, then click **Send**.

![Insert Test Data offered on a selected value in the request URL](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_custom_values_4.png)

Inside a raw body, reference a profile parameter as `@|ParameterName|` and a data generator as `!|Number.digits(int:3)|`. Unicode values work in every field. Paste them in.

## Step and test case settings

Every setting that changes how a test case runs, at the step level and at the test case level: timeouts, retries, screenshots, prerequisites, data-driven runs, and the workflow fields.

### Step options

Click a step to reach its inline controls, or the ellipsis icon `⋮` for the rest. The options behave the same across application types.

- **Clear Step**: empties the step's action and data. The eraser icon does the same
- **Clone Step**: duplicates the step
- **Delete Step**: removes the step
- **Step Above** and **Step Below**: insert a neighbor. **+ Add new step** appends at the end
- **Disable Step**: skips the step at run time without deleting it
- **Ignore Step Result**: keeps the step's outcome out of the test case result
- **Enable Visual Testing**: adds a visual comparison on the step

To reorder steps, drag the `⋮⋮` handle, then click **Save New Order**. In the recorder the button reads **Save Order**.

In the recorder, clicking an element in a step offers **Edit**, **Change**, and **Create Element**. **Pause** and **Stop** control the session.

### Step settings

Click the ellipsis icon `⋮` and select **Step Settings** for the full per-step configuration.

- **Max. wait time**: fails the step past the limit. The maximum is 120 seconds
- **Retries on step failure**: re-attempts a failing step, up to 10 times
- **Screenshot capture**: Always, Only on step failure, No screenshot required, or Use step level settings. This applies to execution only
- **Pre-Requisite**: a step in the same test case that must pass first
- **Stop Test Case execution on Test Step**: stops the test case when this step fails. On by default
- **Ignore this step result in Test Case Result**: excludes the step from the test case result
- **Disable Step**: skips the step. Off by default
- **Enable Visual Testing for the Step**: captures and compares the UI on this step
- **Enable Accessibility Testing for the step**: validates the step against accessibility standards
- **Highlight element in screenshot**: marks the acted-on element in captures

**Highlight element in screenshot** is off until Testsigma support turns it on for your account. It then appears under **Settings > Preferences**.

![The Step Settings panel, with the timeout, retries, screenshot, and per-step toggles](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_step_settings_1.png)

**Global Step Timeout**, under **Additional settings** in the Ad-hoc Run overlay, can override these values. With the toggle off, the global value applies only to steps without a custom timeout. With it on, the global value overrides every step.

### Change several steps at once

1. Hover over a step number to turn it into a checkbox, then select the steps. **Select All** selects every step.

2. Select **Update Settings**, **Create Block**, **Create Step Group**, or **Delete** from the menu bar.

3. Click **Exit Bulk Action** to leave the mode.

![Bulk action mode, with Select All, Update Settings, Create Block, Create Step Group, and Delete](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/update_bulkactions_3.png)

Bulk **Update Settings** cannot change **Pre-Requisite** or the retry count. Set those on each step.

### Test case settings

The right-side utility panel holds everything that belongs to the test case rather than to a step.

- **Test Case Info**: rename the test case, edit the description, and read the created and updated timestamps
- **Ad-Hoc Runs**: the test case's ad-hoc run history
- **Activity**: history and comments
- **Help**: examples, the action list, and get-started guides

**Test Case Settings** configures how the test case runs.

- **Pre-Requisites**: another test case that must run first. **Always run Pre-requisite** executes it on every run. **Only execute failed Pre-requisite iteration(s)** reruns only the failed iterations of a data-driven prerequisite
- **Test Data Profile** and **Test Data Set**: bind the test case to a profile and select the set
- **Data-Driven**: runs the test case once per data set, filtered by Iteration (greater than, less than, between), Set Name (equals, contains, starts with, ends with, between), or Parameter
- **Fail Test Case if Visual Testing Fails**: a visual mismatch fails the test case
- **After Test Case**: cleanup steps that run after the test case, with **Fail the Test Case** or **Show Test Case Result** when they fail. **Mark this for AfterTest Suite** runs the test case in a suite's cleanup phase

![The Test Case Settings panel, with prerequisites, test data profile, and the data-driven filters](https://s3.amazonaws.com/website-static-docs.testsigma.com/new_images/projects/Updated_Doc_Images/Test_Case_Settings_New.png)

**Manage Test Case** carries the workflow fields.

- **Status**: Draft, Review, Ready, Obsolete, or Rework
- **Priority**: Critical, Major, Medium, or Minor
- **Assignee**: notified when the test case fails during a review
- **Reviewer**: the reviewer for the test case
- **Test Type**: Unit Test, Integration, Functional, Non-functional, or User Experience
- **Requirement**: the linked requirement, for traceability
- **Labels**: the labels attached to the test case

Custom workflow fields are defined under **Custom fields**.