Commands reference

Commands are the building blocks of an Agent. They run top-down, and certain commands (repeaters, lists) loop their children. The root of every command tree is the Agent command, whose properties set Agent-wide behavior: browser engine, input parameters, proxies, identity, success criteria, and runtime data.

​

Common to every command

The per-command pages below document only what is specific to each command. Every command shares the same basics, documented once here:

• Options and Text tabs — every command’s configuration dialog has an Options tab for its properties and a Text tab that jumps to the command’s position in the Agent text.
• General command settings — every command has a Command name, an optional Comments field (internal documentation only, never exported), and a Disabled checkbox that makes the run skip the command.
• Shared property groups — many commands attach the same property sections: Browser properties, Error handling, Export properties, and Files.

To inspect a command’s properties in the Editor, click the options icon in the top-right corner of the command. In most cases the defaults are sufficient — adjust them when a site’s behavior demands it.

​

Browser

Navigation and page loading — how the Agent moves through the web and reaches the content it needs.

​

Actions

Simulated user interactions — clicking, typing, waiting — for dynamic and interactive pages.

​

Extract

The heart of an Agent — pulling text, files, and structured records out of pages.

​

Pagination

Two ways to walk multi-page results: use Action repeater when the site pages via JavaScript events, and Link repeater when pagination buttons are real URL-backed links.

​

Control flow

Utility commands for branching, error handling, and custom logic.

​

Export

Where extracted records go — file formats and destinations.