Skip to main content

Inspect and Record

Automating a screen means telling a node which element to act on. Typing selectors by hand is slow and brittle, so Robomotion has you point at the thing instead.

Two features, both provided by the Robomotion RPA browser extension and the Universal Inspector:

What it doesUse when
InspectPoint at one element; its selector goes into the node you were editingYou are building a flow and need one more selector
RecordDo the task once; Robomotion writes the nodes for youStarting a new automation, or capturing a long sequence

Recording is the faster way to begin and inspecting is the way to refine, so most people use both: record the walkthrough, then inspect to fix up the one step the recorder guessed wrong.

It is not only browsers

The feature is called Browser Inspect for historical reasons; the inspector is universal. Which one you get is decided by the node you are inspecting from:

Node namespaceInspector
Core.Browser.*, Robomotion.WebAutomation.*Web
Robomotion.WindowsAutomation.*Windows
Core.Linux.*Linux
Robomotion.MacAutomation.*macOS
Robomotion.JavaAutomation.*Java
Robomotion.ImageAutomation.*Image

Each has both an inspect and a record mode. The web one runs in the browser extension; the desktop ones run through an agent on the machine you are automating, which is why the Record dialog asks you to select a machine and says "No Connected Agents" when there is none.

Setting it up

Install the extensionRobomotion RPA on the Chrome Web Store. Chrome and Edge are both supported. See Chrome Extension.

For desktop inspection, also install and connect the Robomotion Agent on the machine you want to automate. The extension reaches it through a native messaging host, which the agent installer registers.

Design time only

The extension is needed while you are building a flow. A robot running a published flow does not need it — by then the selectors are already in the nodes.