|This content applies solely to Appian RPA, which must be purchased separately from the Appian base platform.|
Automation has the potential to speed up routine tasks in your business, but getting started comes with its own learning curve. The Appian RPA task recorder tool makes the most of your development time. The tool observes and captures your interactions onscreen and applies them in a robotic task definition. Rather than manually creating each robotic task action, the robotic task definition is automatically set up to match your actions. Just use the browser or application as you normally would, and the tool does the rest!
This page describes the task recorder tool, its capabilities, and how to use it to efficiently develop robotic tasks.
The task recorder is a built-in tool that records your interactions in a web browser or Windows application and translates them into actions in a robotic task definition. You don't need to spend time finding the exact selector to identify the element on the screen to interact with. Simply click on that part of the screen, and the task recorder detects it for you. You can also use the task recorder to extract data and save it to a robotic task variable.
The task recorder is available for Windows applications as well as web browsers including Chrome, Firefox, and Edge (including IE Mode). Other browsers aren't yet available for the task recorder.
The task recorder's Windows automation capabilities do not work with JavaSwing apps.
When used with a browser, the task recorder needs the appropriate web driver to launch properly. Appian can automatically update the web driver, or you can add it manually as a support file in your robotic task.
If you're using Edge (IE Mode) with the task recorder, you'll need to complete additional setup steps.
When you use the task recorder to capture interactions in a section of the robotic task definition, a new action group is created. When you begin recording within an action group, the new actions are added to the end of that group.
Use the task recorder to capture actions within Windows applications and web browsers.
For Windows automation, you can use the Interact with element action, with some limitations on interactions for certain control types:
|Control type||Supported interaction options||Unsupported interaction options|
|ComboBox||Collapse, Expand||Set value, Focus|
|Edit||Set value||Update value with username, Update value with password|
|Menu item||Click, Expand||Collapse|
|Radio button||Select||Add to selection, Remove from selection|
|Tree item||Select, Expand, Collapse||Click|
|Window||n/a||Close, Focus, Maximize, Minimize, Restore down|
If you need to use the unsupported interaction options in your robotic task, you can manually configure them in the Interact with element action.
The following Browser actions are available in the task recorder:
Other modules and actions aren't yet available in the task recorder. To take advantage of additional functionality, use the appropriate module when you build a robotic task definition manually.
You'll start the task recorder in the robotic task definition.
To start recording:
In the empty section where you want to record, click START RECORDING. If you're recording actions in a section that contains actions, start the task recorder in the section toolbar.
The task recorder opens on the host machine, where you can choose the application to record. Switch over to your host machine, whether it's a virtual machine or separate computer, to get started.
After you launch the task recorder, you'll choose where to begin recording actions.
Select an application where you want to record your interactions.
Applications are categorized as CURRENTLY OPENED if they are running on the host machine. Closed but available applications appear in the ALL list. Applications shortcuts must be in the Windows start menu for the application to appear in the list.
You can only record in one browser at a time. To open a different browser, you'll need to close the open browser first.
When you interact with the screen, such as clicking a link or button, typing in a text field, or selecting a value from a dropdown, a new action is recorded. The task recorder captures the information that the robotic task needs to replicate the action, such as selectors, interaction types, and attributes. The task recorder names each action based on the action type. Click on the action's name to edit it. Your change will be saved when you finish recording.
The following table shows the configuration options that appear with supported actions:
|Web browsers||Get attribute||Interact with, Attribute, Save into|
|Windows applications and web browsers||Interact with element||Interaction|
|Web browser||Navigate to URL||URL|
|Web browser||Open browser||Browser, URL|
|Web browser||Select frame||Switch to, Interact with (when applicable)|
|Web browser||Switch tab or browser||Title|
|Web browser||Wait for attribute||Interact with, Attribute, Attribute value|
The options are automatically configured when you record. You can edit any of the options to manually configure them:
Many actions will involve interacting with the screen in some way. The task recorder captures these interactions. The available options include:
For example, if you click into a text field and type something, the action uses Update value as the Interaction and uses your text input in the Value field. Similarly, if you click into a dropdown menu and select an item, the action uses Select dropdown and logs the item in the Select By field. Other interactions captured include clicking on a link or button, selecting or deselecting a checkbox, deselecting a dropdown option, and clearing a value.
For elements where the interaction doesn't require a click, you can hover on an element to reveal additional options. These interactions include Get Attribute and Wait for Attribute.
Usually, when you input text on the screen while recording, the task recorder saves the value as plain text. This could present a security issue if this information is sensitive, like login credentials. To prevent security issues, task recorder detects when you're entering a username or password and guides you to use credentials instead.
If part of your robotic task requires you to log in to a web site, the task recorder uses Update value with username or Update value with password for the Interaction type. The recording clears the values associated with these actions when you finish recording.
As you proceed, confirm that the right interaction type is selected. If the task recorder doesn't properly detect that you entered a username or password, you can update the interaction type.
Once you're done recording, you'll need to edit the applicable actions in the robotic task definition to use credentials:
Learn more about credentials.
The Get attribute and Wait for attribute actions include the option to save data found within an element. To select a target for this data, choose from the list of robotic task variables or create a new one. The robotic task stores this data in the variable you choose.
To create a new robotic task variable in the task recorder:
In the Task Recorder window, find the Save into field. This field appears for the actions listed above.
Select the Parameter, Required, or Multiple checkboxes to configure your new variable with these specifications.
When you finish recording and return to the robotic task definition, the variable is available to use throughout the robotic task. To learn more about how variables pass data in a robotic task, see Using Robotic Task Variables.
To save your progress and finish recording:
In the RPA console, you'll see the actions available in the robotic task definition area. Each action you recorded corresponds to a newly added action in the robotic task definition. If you see something you want to change, you can manually edit these actions as you normally would.
For example, you may want to edit the recorded actions to:
To exit the task recorder without saving your actions, close the task recorder window.
Edit an action in the task recorder to adjust any of the fields that were automatically configured. The options vary based on the action type. The task recorder names each action based on the action type. Click on the action's name to edit it. Your change will be saved when you finish recording.
You can delete actions as you record. Click the trash icon to remove unnecessary actions, such as mistaken clicks, to clean up your recording before generating the robotic task definition actions.
The task recorder automatically populates the element selector when you interact with that element on the screen. You can edit this value if you prefer to use a different selector.
To help you confirm your updated selector points to the correct element, the task recorder highlights the element. You don't have to leave the task recorder or go through additional steps to test if the edited selector is still pointing to the element you intend to.
To edit the selector:
If you are recording in a browser other than Chrome or Edge, the element will be highlighted when you move your cursor to the Interact with field and edit the selector type or value.
The task recorder highlights matching elements that are visible on the screen. As you interact with the screen, keep in mind that these visible elements may change.
At any point during your recording, you can return to previously recorded actions to review the action configuration, including selectors. Keep in mind that the task recorder checks the current browser window of the active application. If you are reviewing a previously recorded action, the task recorder may indicate that no matching element is found if the current window isn't the proper context for that action.
To verify the action's selector is correct, return to the browser window that matches the context for the configured action. The task recorder will highlight the visible elements that match the selector.
Example: Consider your interaction with a search engine. You visit the engine, type in a term, click a search button, view results, and click the top link. The web page changes between these actions, and so do the visible elements. When you record this interaction, the first action includes elements such as a search field and button, while later actions may only include text links and descriptions.
If you're viewing the search results page, but you're editing the action to input a search term, you may see an error message indicating that no elements match the selector. This is because the current page doesn't have an element (the search field) that matches the configured selector.
To confirm the selector is configured correctly, return to the page where the search field is present. The task recorder highlights that element now because it is visible on the current browser window.
Edge (IE mode) is supported when interacting with websites designed to support this browser. Some websites may specify legacy document modes in their source code, which could make it incompatible to use with the task recorder. See Microsoft's documentation for more information.
You may see an unreliable connection warning message in the task recorder when using Edge (IE Mode). If you want to record your interactions in Edge, you'll need to complete some additional steps before you begin:
3.150version of the web driver is required for reliable performance.
There are some limitations to using the task recorder with Edge (IE Mode):
You'll need to repeat these steps on each host machine where you plan to record using Edge (IE Mode).
The first time you launch task recorder in Edge (IE Mode), you'll see a message prompting you to install a user certificate. The task recorder uses this user certificate to help establish a reliable connection with the browser, and ensure that your steps are recorded correctly.
To capture and extract the developer's interactions with the browser, task recorder establishes a WebSocket connection with the browser. Whereas all other browsers support the standard WebSocket protocol (WS), Edge (IE Mode) only supports the secure WebSocket protocol (WSS).
If the connection is not established, actions can still be recorded. However, some specific steps, such a click that navigates to a different page, won't be recorded correctly.
In order to establish a WSS connection with Edge (IE Mode), you need to manually install a certificate on the host machine. The Appian RPA agent generates the certificate to create the connection. The certificate is called
appian-rpa-recorder-localhost.crt, and is stored in the
After you complete the following steps, the certificate remains on this host machine. As long as the host machine remains as an entry in the RPA console, you won't need to install another certificate when you record your interactions in Edge (IE Mode) on the same host machine. However, if you remove the host machine and re-add it at a later time, you'll need to reinstall the certificate.
To install the certificate:
In the Certificate window, click Install Certificate.
A Security Warning appears to show you that this certificate will only be used for localhost connections (
127.0.0.1). Click Yes.
To verify the certificate was installed properly:
127.0.0.1should appear as a certificate in this folder.
To help keep you on track, the task recorder may display messages to show you when it's unable to do something.
|Error opening browser||This message appears if the task recorder was unable to open the browser you selected.||Ensure no other browsers are open and try again.|
|Error opening application||This message appears if the task recorder was unable to open the application you selected.||Open the application on your desktop, return to the task recorder, and use the Select Application menu to try again.|
|Error connecting to application||This message appears if the task recorder was unable to locate the application you selected, even if it is open.||If the application is not yet open: open the application on your desktop, return to the task recorder, and use the Select Application menu to try again.
If the application is open: restart the application on your desktop, return to the task recorder, and use the Select Application menu to try again.
|Unreliable connection||This message appears if you navigate to a site that violates your network's content security policy. While actions can still be recorded on the page, some steps may not be recorded correctly.||To reliably record actions in Edge (IE) mode browsers, install the user certificate. If issues persist, you may need to contact your network administrator.
To reliably record actions in other browsers, you may need to contact your network administrator to adapt security configurations.
On This Page