Prompt
A prompt is the input a large language model uses to generate text. Prompt engineering is the practice of writing effective instructions so a model consistently produces content that meets requirements.
On this page
Source content checked . English translation published 2026-10-27.
In business automation
A good prompt defines the task, input, and desired output instead of merely asking the model to act as an expert. In an automation, the output must fit the next step, and the instructions should make clear what to do when information is missing.
Examples
These examples illustrate a possible workflow. Adapt the connections and rules to your own systems.
Business process example
An email is sent to a model with a request to return a category from a list and a one-sentence summary. The instruction says to flag missing information instead of inventing an order number, and the system validates the structure before continuing.
Relevant tools
- Dify
- Instructions and variables can be configured in language steps, and the output can be connected to the workflow.
What to check
- Test prompts with ambiguous and adversarial input. Careful wording is not an authorization mechanism and does not guarantee valid output.