nodes / extension
Get Accessibility Tree
Get visible interactive and semantic elements from the opened page
List the buttons, links, form fields and other controls you can see on the active tab, with a name and a selector for each.
What it does
Section titled “What it does”Get Accessibility Tree scans the active tab (or the part inside DOM Selector) for elements a person can interact with: links, buttons, inputs, lists, text areas, expandable sections, and anything the page marks with a role. It keeps only the ones that are visible on the page, and returns one item per element.
For each element you get:
role: what it is, such aslink,button,textbox,checkbox,comboboxorslider.name: the label a screen reader would use (its label, title or text).text: its visible text.selector: a selector you can pass to other page nodes to act on it.disabled, and when relevantchecked,selectedandexpanded.boundingBox: its position and size.
When to use it
Section titled “When to use it”- Give an AI model a compact map of what can be clicked or filled on a page, instead of the full HTML.
- Find the selector of a button or field, then click it with UI Automation or fill it with Form Fill.
- Quickly check a page for unlabeled buttons or links (an empty
name).
Inputs and settings
Section titled “Inputs and settings”| Setting | Type | What it does |
|---|---|---|
| DOM Selector | Page element | The selector of the DOM element to interact with. |
Leave DOM Selector empty to scan the whole page. To limit the scan to one area (a dialog, a sidebar), click the crosshair button in the field and click that area, or type a CSS selector.
Outputs
Section titled “Outputs”-
Output
Each run returns a list of items with these fields:
| Field | Type | What it holds |
|---|---|---|
| selector | Text | |
| tagName | Text | |
| role | Text | |
| name | Text | The accessible name |
| text | Text | |
| disabled | True/false | |
| checked | True/false | |
| selected | True/false | |
| expanded | True/false | |
| boundingBox | Object |
[ { "selector": "…", "tagName": "…", "role": "…", "name": "…", "text": "…", "disabled": true, "checked": true, "selected": true, "expanded": true, "boundingBox": {} }]Each element is one item, so the next node runs once per element. To give the whole list to an AI node in one go, first combine the items with Aggregate.
Dependencies and credentials
Section titled “Dependencies and credentials”None. This node needs no connection and has no slots for other nodes.
Example workflow
Section titled “Example workflow”- Run the workflow from a Hotkey trigger on the page you want to explore.
- Add Get Accessibility Tree with DOM Selector empty.
- Add Filter to keep items where
{{ $input.role }}equalsbutton. - Combine the buttons into one item with Aggregate, then show it with Display Object to see every button’s name and selector.
Common issues
Section titled “Common issues”- The output is empty. Nothing visible and interactive was found, or DOM Selector matched nothing. Wait for the page to finish loading, or check the selector with Ping.
- A control you can see is missing. It may be inside an embedded frame (not scanned), or built from plain elements without a role, which the page gives no hint about.
- Elements in a closed menu or another tab of the page are missing. Hidden elements are skipped on purpose. Open the menu first, for example with UI Automation.
- The node fails outside the extension. This node runs in the browser extension; install it or run the workflow from the extension.
Related nodes
Section titled “Related nodes”- Get Element Data: full details of one element.
- Get Forms Schema: every field of a form, with its options.
- UI Automation: click, type and select using the selectors found here.