Text Node

A text node is like a blank field placed inside an automation workflow. Just as a form has a blank for writing a name, and what you write there changes the content of the form, what value you put into a text node changes what the connected steps downstream do.

For example, in a workflow that auto-translates thumbnails into different languages, putting the value "French" into a text node lets the AI connected downstream read that value and rewrite its own instructions for French before running. Put "German" in the same spot, and the same workflow produces a German version instead. In other words, you don't need to rebuild the whole workflow—just change the value in this one field to get a different result, creating a reusable structure.

The text node itself doesn't calculate anything or produce a result. It simply holds a value and passes it along to the next step. In more complex automations, where you place these simple input fields determines how easily the whole structure can be reused.

The article explains that "adding a text node that turns the language itself into an input value creates a reusable template where the LLM rewrites the prompt on the fly." Here, a text node isn't some advanced technique—it's just a slot that holds one changeable value in the workflow. Once that slot exists, you don't need to rebuild the workflow for each language; you just swap out the value.

1. In your automation workflow tool, add a new node and set it as a node that holds a text value.
2. Put a value into that node that will change later, such as a language name or phrase.
3. Connect this node to the next node that runs an AI prompt, and write the prompt so it references this value.
4. Run the same workflow multiple times, changing only the text node's value. For example, swap the language value between Spanish, French, and German in turn and check whether the results change accordingly.