Skip to main content

Documentation Index

Fetch the complete documentation index at: https://ardocs.autonome.in/llms.txt

Use this file to discover all available pages before exploring further.

A prompt is a buyer query that AgentRank tracks on your behalf, such as “best hiking boots for wide feet” or “compare trail running shoes under $150”. Your Visibility and competitor metrics are calculated across your active prompts. The better your prompt set reflects real buyer intent, the more meaningful the data becomes.

How prompts are generated

During onboarding, AgentRank generates prompts from your topics using AI. They are designed to reflect the kinds of questions shoppers actually ask before they buy. You review prompts before they go live. You can activate good ones, keep some suggested for later, add custom prompts, or move weak prompts out of rotation.

Intent types

Every prompt has an intent: what the buyer is trying to accomplish. Intent affects the kind of response you get and which brands tend to surface.
IntentDescriptionExample
PurchaseReady to buy, looking for the best option”best trail running shoes to buy right now”
ComparisonEvaluating options side by side”Salomon vs Hoka trail shoes”
Feature-specificResearching a specific attribute or use case”most durable waterproof hiking boots”
Brand checkChecking whether a specific brand is trusted or suitable”are Merrell hiking boots reliable?”
ResearchEarly-stage exploratory shopping”what to look for in trail running shoes”
Purchase and Comparison intents are highest-stakes. Buyers at these stages are closest to a decision, so AI recommendations directly influence conversions.

Prompt states

StateMeaning
ActiveCurrently being tracked
SuggestedGenerated, awaiting your review
InactiveKept in the workspace, but not currently tracked
You can change prompt state at any time from the Prompt Workspace.

Prompt limits by plan

PlanActive prompts
Free5
Starter50
Growth150
CustomCustom
Choose prompts that cover the breadth of your topics and a mix of intent types. A diverse set gives you a much more accurate picture than near-duplicate queries.