How to use the Recipe Extractor.
A recipe extract should separate ingredients and preparation steps from surrounding page content. Preserve quantities, units and source attribution.
Make the workflow fit your task.
Select the actual recipe section, separating ingredients, quantities and ordered preparation steps from commentary. Preserve serving size, timings and notes attached to specific ingredients. Keep the source reference and flag missing instructions instead of inventing a complete recipe.
- What you provide
- Visible recipe page.
- What you get
- Ingredients, quantities, steps and source attribution.
See the input and the result.
Illustrative input and output · a teaching example, not a live WebAct run
Example input
Fictional recipe excerpt at https://example.com/flatbread: yields 2 portions. Ingredients: 200 g flour; 100 ml water. Step 1: Mix flour and water until combined. No subsequent steps supplied.
Completed example
Source: https://example.com/flatbread Yield: 2 portions Ingredients: - 200 g flour - 100 ml water Method: 1. Mix flour and water until combined. Completeness: this excerpt ends after mixing. Cooking instructions are not supplied and have not been invented.
Load this input into the prompt, then copy it to WebAct to try the task. Your result may differ from the illustration.
Decisions and troubleshooting.
Should ingredient quantities be scaled during extraction?
Extract the stated quantities first. Scaling is a separate operation that needs a target serving count and judgment for non-linear ingredients.
Why are ingredients missing when the page appears complete?
The recipe may be collapsed, paginated or loaded separately. Expand the relevant content and check the extracted list against the visible recipe.
Try it with your own source.
Replace the example with your material in the task prompt. Keep the requirements you need, then copy the task into WebAct.
Customize and copy the task ↑