Conditional Prompt
Use AI to decide whether a workflow should keep running. Describe the condition in plain language, and the node stops the workflow branch if the AI decides it is not met.

What It Does
Evaluates your prompt against the current workflow state and returns a true or false decision. If the decision is false, the workflow branch halts immediately rather than continuing to downstream nodes.
When To Use It
Use this to gate a workflow on an AI judgment call, like "continue only if the deal value is above $10,000" or "continue only if the customer sounded frustrated in this call." It is a good fit anywhere you would otherwise need a person to eyeball the data before letting a workflow proceed.
This node has no implicit context. It cannot see the meeting transcript, CRM records, or any prior node's output unless you insert that data into the prompt with a template reference, like {{$triggerNodeId.meetingTranscript}}.
Configuration
Configure the node with these settings:
Field | What it controls |
|---|---|
Prompt | Describe the condition that must be true for the workflow to continue. Required. |
Model | Which LLM evaluates the condition. Choose |
Temperature | Controls consistency of the decision. Lower values (0.0-0.3) make the decision more repeatable. Required, defaults to |
Continue on Failure | When enabled, an error during evaluation lets the workflow continue instead of stopping. This only applies to evaluation errors, not to a deliberate |
Outputs
The node passes these values downstream. Check the Outputs tab on the configured node to confirm exact paths.
Variable | Type | Description | Example |
|---|---|---|---|
|
|
| |
|
|
|
Example Workflow
After a discovery call, check whether pricing was discussed using a Conditional Prompt. If it was not, halt the branch instead of running a pricing follow-up sequence.
Related Nodes
Run Prompt
Filter Prompt
Outcome Validator
Need More Help?
Reach the AskElephant support team by:
clicking the Support tab on the right side of your screen,
emailing [email protected],
or using
@askelephant supportin your dedicated Slack channel.