LLM Mentions Search

Use LLM Mentions Search when you want a workflow step to find AI answers that mention a domain or keyword, then pass those mentions into later workflow steps.

Configuration

Option	Required	Description
Name	No	Label for the step in the workflow canvas.
Target	Yes	Choose Keyword or Domain.
Keyword / Domain	Yes	The target value. This field supports workflow variables through Insert Variable. Keywords can be up to 250 characters. Domains can be up to 63 characters.
Search scope	No	Limits where the target is matched. Keyword scopes include question, answer, brand entities, and fan-out queries. Domain scopes include sources. Search results is available for ChatGPT domain searches only.
Match type	No	For keywords, choose Word match or Partial match.
Include subdomains	No	For domains, include mentions of subdomains as well as the root domain.
Platform	Yes	Search Google or ChatGPT mentions. ChatGPT locks Language to English and Location to the United States.
Language	No	Language value in Language. You can choose from the list or enable Custom Value. ChatGPT mentions always use English.
Location	No	Numeric location code in Location (for example `2840`). Choose from the list, or enable Custom Value with a numeric code or a variable that resolves to one. ChatGPT mentions always use the United States.
Filters	No	Up to 8 optional filter rows joined by AND or OR. Available fields include platform, location code, language code, AI search volume, first response at, last response at, and model name.
Sort by	No	Up to 3 sort rules.
Number of results	No	Number of mentions to return, from 10 to 100.
Offset	No	Skip this many mentions before returning results.
When the step fails	No	Controls whether the workflow should Terminate Workflow or Continue if this step fails.

Add this step from the Research group in Search steps…. The target field supports Insert Variable. The Language and Location controls can use the built-in lists or switch to Custom Value, except when Platform is ChatGPT.

Output

Click Run in the step header to test the step. Fetch Hive shows the matching mentions in Output after the run completes. Use the variable picker in a later step to insert the exact output path available for that run. The base reference is:

{{STEP_IDENTIFIER.output}}

The output is an array of mention objects. Each item includes the question, answer, platform, model, location, language, search volume, sources, and related entities. To use the first answer:

{{step_2.output[0].answer}}

Use the variable picker after a test run to inspect nested fields such as sources and brand_entities.

Example

Add LLM Mentions Search from the Research group in Search steps…. Set Name to something like Brand mentions. Choose Keyword and enter the brand or phrase in Keyword. If the target should reuse data from earlier workflow steps, click Insert Variable and add that reference. Set Platform to Google or ChatGPT. Add filters or sort rules if you want a narrower page of results. Click Run and review the returned mentions in Output before using them in a later step.

Notes

ChatGPT mentions are available for English in the United States only. The editor disables Language and Location when ChatGPT is selected.
You can add up to 8 filters and 3 sort rules.
Each run returns a single page of up to 100 results. Use Offset to skip earlier mentions.
Hosted runs convert the recorded service cost plus a 6% service fee into task credits. See Task costs.

Getting Started

Live Chat

Prompts

Workflows

Hive Agents

Agents

Knowledge Bases

Experiments

Your Workspace

Monitoring

User Tracking

Resources

Configuration

Output

Example

Notes

​Configuration

​Output

​Example

​Notes

Configuration

Output

Example

Notes