AI Prompt
The AI Prompt activity sends a single prompt to a large language model (LLM) and returns the model’s text response. Use the activity for fast, direct artificial intelligence (AI) completions that do not require workflow tools.
Before you begin
The AI Prompt activity requires an LLM agent endpoint target. Confirm the following before adding the activity to a workflow:
-
A target of type AI Agent Endpoint is configured and assigned to the workflow.
-
At least one model is available on the target.
-
The prompt content is available, either as static text or as a workflow variable.
Inputs
The activity groups inputs into a Request section for everyday use and an Advanced section for response tuning. The following tables describe each field.
Request inputs
|
Field |
Required |
Description |
|
Model name |
Yes |
The model that runs the prompt. The selected value always overrides the model stored on the target, so the activity runs with the model chosen here. |
|
System instructions |
No |
Sets the AI’s role and expertise. Shapes the tone, focus, and constraints applied to every response. |
|
User prompt |
Yes |
The question or task sent to the AI. Supports workflow variables, so the field can reference form inputs or incident data. |
Advanced inputs
|
Field |
Required |
Default |
Description |
|
Temperature (chat response variability) |
No |
1.0 |
Controls creativity versus consistency on a scale of 0.0 to 2.0. A value of 0 returns deterministic output, and a value of 2 returns the most creative output. |
|
Max output tokens |
No |
None |
Limits the length of the AI response. Higher values increase response time and cost. |
System instructions and the user prompt accept workflow variables. Reference a variable to pass dynamic data, such as a form field or an incident description, into the prompt at run time.
Example
The following example configures the activity to summarize an incident description that a previous step stored in a workflow variable.
-
In Model name, select a model from the target, such as gpt-4o.
-
In System instructions, enter You are a network operations assistant. Summarize incidents in two sentences.
-
In User prompt, reference the variable that holds the incident text, such as incident.description.
-
In the Advanced section, set Temperature to 0.2 for consistent summaries.
-
Save the activity and run the workflow.
The activity returns a concise summary in the content output field, which later steps can reference.
Outputs
The activity returns the model response and token usage. The following table describes each output field.
|
Field |
Type |
Description |
|
Content |
String |
The text response generated by the model. |
|
Usage > Input tokens |
Integer |
The number of tokens in the prompt sent to the model. |
|
Usage > Output tokens |
Integer |
The number of tokens in the model response. |
|
Usage > Total tokens |
Integer |
The combined input and output token count for the request. |
Next steps
-
To orchestrate workflows or call tools from a model, use the AI Agent activity.
-
Reference the content output in a later activity to act on the model response.
-
Adjust Temperature and Max output tokens to balance response quality, speed, and cost.