Prompts

Defined questions run in ChatGPT at scale to quickly find mentions and citations of your brand and competitors.

Prompts are the questions your customers ask AI models. Airefs runs them in ChatGPT with web search enabled — the only mode that returns real citations — and records the full response: which brands were mentioned, at what position, and which URLs were cited. You can add prompts manually or pick from Airefs' suggested list, tag them to group by topic, and filter by country to track specific markets. The Brand mentioned smart filter surfaces prompts where your brand recently gained mentions — a fast way to spot where your AEO work is starting to pay off.

The prompts table

Airefs Active prompts table showing prompt text, country flag, LLM icon, rank position, brand mention logos, citations, and last run date
Active prompts table — rank, mentions, and citations at a glance across all running prompts

The Active tab shows all your running prompts with the following columns:

  • Country — the market this prompt runs in
  • LLM — the AI model used (ChatGPT by default)
  • Rank — the position your brand appears in the response, with an arrow indicating whether it's trending up or down since the last run
  • Mentions — brand logos for every brand mentioned in the answer, so you can see at a glance who appears alongside you
  • Citations — URLs your content was cited in the response
  • Last Run — the date of the most recent execution

At the top of the list, a summary line shows your overall mention rate and citation rate against your tracked competitors — a quick read on where you stand across all active prompts.

Airefs prompt execution view showing the full ChatGPT answer with brand mentions and citations listed in the side panel
Prompt execution view — full ChatGPT response with every mention and citation

Drill into any prompt to see the full AI response with every mention and citation. You see which brand appeared at #1, which sources were cited, and how results shift over time. The Rank column shows which position your brand appears in the response — being the first brand mentioned in an answer is very different from being fifth.

Suggested prompts

Airefs Suggested prompts tab showing three columns: Solution research, Alternative research, and Education research with pre-built prompts ready to add
Suggested prompts tab — 25 pre-built prompts across three research categories, one click to activate

The Suggested tab gives you a pre-built list of prompts organised into three research categories. You can add any of them to your active tracking with one click.

Solution research

Buying-intent queries from people actively evaluating tools in your category — "best sales automation tools for startups", "top CRM solutions for sales teams", "best marketing software for small businesses". These are the highest-value prompts: appearing here means appearing in front of buyers who are ready to make a decision.

Alternative research

Competitor comparison queries — "best alternatives to [competitor]", "solutions similar to [competitor]", "what tools are similar to [competitor]". These capture buyers who are already using a competitor and looking to switch. Being present here is one of the fastest ways to gain share from competitors who already have market awareness.

Education research

Informational queries your target audience asks about the industry — "how does AI-powered CRM work?", "what are the benefits of integrating marketing and sales?", "best practices for using AI in marketing". With these prompts, you likely won't be mentioned directly — brand mentions are less common in educational answers. Your goal is to appear in the Citations column, meaning ChatGPT used your content to formulate the answer. That's a powerful signal and a path to growing brand presence over time.

Web search must be enabled on prompts to get citations. Airefs enables this by default — without it, ChatGPT draws only from its training data and returns no source URLs.

Prompt limits by plan

Plan Prompts
Lite25
Pro60
Expert150