Extract
Pulls the fields you describe out of free text into new columns, such as company names, dates or amounts from email bodies.
Also known as: Entity extraction
How it works
Describe in the prompt box which text column to read, then add one field per column you want, each with a field name and a description (press Add field for more). Press Generate and the app asks the AI service for Python, spaCy or regex based, that pulls those fields out of the text into new columns. The code lands in the Code tab, where you can read and edit it before it runs. Generate is enabled once the prompt is filled in; the panel flags an empty field list as missing until code exists, and the node has nothing to run until then. A Save as custom node button in the panel header keeps the result in your library.
Inputs and outputs
Inputs
data— Table
Outputs
data— Table
Settings
The Attributes panel of Extract, as in the app: the controls work, but nothing runs — there is no data behind it.
Example
Email bodies mention a company, a date and an amount. Prompt: "Read the body column." Fields: company ("the sender's company name"), invoice_date ("the date mentioned") and amount ("the money amount as a number"). After generating and reviewing, run it and use Cast Types on amount if it comes back as text.