AI is answering your buyers right now. This is your scoreboard.

Prompt sets run on a schedule across ChatGPT, Claude and Gemini, plus Perplexity, Google AI Overviews and other engines you add. Every response is scored — were you mentioned, were you cited, how were you framed — so AI visibility becomes a trend you manage, not an anecdote.

Free · No credit card · Your plan in minutes

NNorthwind CeramicsConnect your agent

AI visibility

Example
VisibilityCompetitorsGaps
–

Visibility score

Asking ChatGPT, Claude and Gemini…

Kiln & Co is named in 5 of 9

  • ChatGPT–
  • Claude–
  • Gemini–

Your tracked prompts: who each answer named

What’s the best handmade stoneware for everyday use?

Which ceramic mugs are dishwasher safe and actually last?

  1. 1Your buyers' questions go to ChatGPT, Claude and Gemini
  2. 2Each answer shows who it named: you or a rival
  3. 3Your score: 41 out of 100
  4. 4The brands named and the sources they cite
  5. 5What to fix this week

AI visibility tracking

How it works

  • Prompt sets built from your market

    The generator proposes the questions your buyers actually ask — comparisons, alternatives, problem-aware, branded — classified by type, steerable with plain-language guidance, and editable prompt by prompt.

  • Scored responses, not screenshots

    Each engine’s response is captured and scored on a rubric: visibility, depth, framing, citations, competitive position. Mention rate and citation rate aggregate per round, so movement is measurable.

  • Competitors in the same frame

    Tracked responses record every brand mentioned, so you see who AI recommends for your category today, which competitors keep surfacing, and whether your share of the answer is growing.

  • Connected to the agent that can act

    The scoreboard lives next to an agent that ships fixes — answer-first content, structure, schema — behind approvals. Measurement and remediation are one loop, not two tools.

Buyers increasingly ask AI instead of searching. Whether you appear in those answers is now a marketing metric — and like any metric, it only improves once something measures it continuously.

FAQ

Common questions

Which engines are tracked?
ChatGPT, Claude and Gemini by default, with Perplexity, Google AI Overviews and other engines one tick away. Prompt sets run against each on your cadence, and per-engine scores make response-style differences visible instead of misleading.
What exactly gets scored?
Each response is scored on a rubric covering visibility, depth, framing, citation, and competitive position, plus flat mention and citation flags — aggregated into per-round and per-engine trends.
How do I know which prompts to track?
The generator analyzes your site and proposes up to 25 prompts across categories, with plain-language steering (e.g. “mostly unbranded prompts”) and per-prompt editing before anything runs.
Can it improve the scores, not just report them?
Yes — the same agent ships the remediation: answer-first pages, FAQ and schema coverage, structural fixes, all behind approvals. The tracking then shows whether the needle actually moved.

See it on your own site first

The free audit grades your site and shows the plan Magister would execute — the fastest way to see what these capabilities do for you.

Run the free audit

Free · No credit card · Your plan in minutes