Quickstart

This walkthrough builds a tiny scraper: open a page, pull a quote out of it, and log the result. You'll touch every core part of Orchestra on the way — the palette, the embedded browser, selectors, variables and the run controls.

1. Open a page#

Launch Orchestra. The center panel is a real embedded browser. Type quotes.toscrape.com into its address bar and press Enter — this site is made for scraping practice.

Browsing by hand is fine for exploring, but a flow needs its own Navigate step so it always starts from the right page:

1. In the left palette, find Navigate (Browser category) and click it. A step appears in the flow panel.
2. Select the step and set its URL to https://quotes.toscrape.com in the config panel on the right.

2. Extract some text#

1. Add an Extract step from the Data category.
2. In the config panel, click Pick selector. The embedded browser enters picking mode — hover highlights elements.
3. Click the first quote on the page. Orchestra fills in the most robust selector it can find (a popover offers alternatives if you want a different one).
4. Give the extracted value a variable name: quote.

3. Log the result#

Add an Output step (Data category) and set its value to $quote. Anything you write with a $ prefix reads a variable.

4. Run it#

Press Start in the top-right. The flow runs top to bottom: each step lights up as it executes, the browser navigates, and when it finishes the Output tab of the bottom panel shows your quote.

That's the whole loop: add steps, point them at the page, run, read the output.

Prefer not to build by hand?#

Click Orchestrate in the browser toolbar and just use the page — Orchestra records your clicks, typing and navigation as steps automatically. See Recording.

Where to go next#

• The interface — what every panel does.
• Selectors — how Orchestra finds elements, and how to pick good ones.
• Running flows — stepping, pausing, and running from a specific step.
• Output & data — exporting scraped data to JSON or CSV.