Websites, brand and AI visibility
The RankX AI MCP tools that find your Website, read its brand profile and competitors, manage tracked prompts, and report AI visibility and AI Shopping results.
These tools let an assistant find your Website in RankX AI, read who the
business is and who it competes with, manage the prompts RankX AI checks, and
report how often AI assistants name your brand and recommend your products.
Every RankX AI session starts here, because list_projects returns the Website
id almost every other tool takes.
| Tool | Scope | What it does |
|---|---|---|
create_prompt | write | Adds an AI visibility prompt to a Website's tracked set. It can be an AI visibility prompt or a shopping prompt. It is active immediately, so it joins the schedule and counts against the plan's per-Website prompt cap. |
get_brand_profile | read | The Website's brand profile: who the business is, where it operates, its audience, its positioning and its services. The location fields here are the authoritative ones. |
get_product_shopping_detail | read | One product: the shopper questions it appears on, the merchants that shared those answers, and the cards the surface actually showed. |
get_project | read | One Website by id. Note the explicit-market flag: when it is false, the country and currency are defaults rather than facts about the business. |
get_prompt_visibility | read | Brand mention rates across the AI assistants RankX AI tracks for the Website, per assistant, over a chosen window. Rates divide by checks with a known verdict, so a null rate means unknown rather than zero. |
get_shopping_prompt_detail | read | One shopping question in detail: how often its answers showed a product carousel, whether the Website's monitored products were in those carousels and at what position, and which merchants took the slots instead. Every rate carries its own denominator. Free: it reads answers already collected. |
get_shopping_visibility | read | Whether a Website's products are recommended when a shopper asks an AI assistant, per surface, over the last 28 days. Every figure carries its denominator, and a null means no answer showed products at all. |
list_competitors | read | The Website's saved competitors with their domains and aliases, which is the set RankX AI compares against in visibility and ranking reports. |
list_projects | read | Lists the Websites in the workspace. Call this first: every other Website-scoped tool takes an id returned here. |
list_prompt_candidates | read | The candidate prompts an earlier research_prompts run found, best first. Free, so it is the call to make before spending on fresh research, and a candidate becomes a tracked prompt with create_prompt. |
list_prompts | read | The AI visibility prompts tracked for a Website, which are the questions RankX AI asks the assistants on your behalf. |
list_shopping_competitors | read | The merchants whose products were recommended instead of yours. It is built from product cards that did not match your catalogue, so an incomplete catalogue inflates the list. |
research_prompts | spend | Discovers the questions people ask AI assistants in a market, with an estimated demand figure and a relevance score for the Website, from seed topics or from the brand profile. The demand figure is modelled from Google's People Also Ask results and is not a count of ChatGPT or Gemini queries, and that caveat travels in the response rather than only in this documentation. It spends only on a fresh run: repeating a recent search is served from cache at no cost. |
run_prompt_check | spend | Runs a live AI visibility check for one tracked prompt across every assistant enabled for the Website. Priced per assistant checked, and a failed assistant is refunded. |
run_shopping_check | spend | Checks one shopping question now, outside its schedule. Priced per enabled shopping surface, and it runs in the background: the result is read afterwards with get_shopping_prompt_detail. One question can be checked by hand once every five minutes. |
set_prompt_status | write | Pauses or resumes a tracked prompt. Only active prompts run on a schedule, so this is the reversible way to stop one costing credits. Nothing is deleted. |
The table above is generated from RankX AI itself rather than written by hand, as at 30 September 2026 (source b27dec96). When the product changes, this page changes with it.
A typical request to an assistant with these tools connected:
List my websites, then show me AI visibility for the first one over 30 days, broken down by assistant, with the number of checks behind each rate.
Resolving the Website
list_projects is the entry point, and RankX AI instructs every connected
assistant to call it before anything else. Every other tool here takes the
projectId it returns.
Do not guess an id. An id outside your credential's scope answers "not found" rather than saying whether it exists, so a guess produces a confusing error.
get_project carries one field worth reading: whether the customer explicitly
chose a market. When they did not, the country and currency on the Website are
defaults, and an assistant should not report them as the market the business
trades in. get_brand_profile holds the authoritative location facts instead,
along with the audience, positioning and services.
Reading AI visibility
get_prompt_visibility returns a brand mention rate per assistant over a 7,
30 or 90 day window, with the number of checks behind each rate. Three rules
change how the answer should be read.
The rate divides by checks with a known verdict. A check that ran but could not be analysed is left out of both halves of the fraction and counted separately.
If nothing has a known verdict, the rate is null, which means
unknown. Reporting it as 0% would tell the customer they are invisible when
nothing has been measured.
Per assistant is the real figure. The assistants switched on for a Website answer differently by design. A 40% rate on one and 10% on another is two findings, not an average of 25%.
One run is not a measurement. Assistants are non-deterministic, and two runs of the same prompt often name different brands. The reliable unit is a rate across a panel of prompts over several runs.
Managing the prompt panel
list_prompts reads the panel, create_prompt adds to it, and
set_prompt_status pauses and resumes. A new prompt is active at once and joins
the scheduled visibility run.
A prompt is a recurring cost. Every active prompt is checked on every
assistant enabled for the Website, on every scheduled run, and each check is
charged. So adding a prompt is a spending decision, even though create_prompt
carries the write scope rather than spend.
Pausing is the reversible way to stop that cost. set_prompt_status moves a
prompt to inactive, which takes it out of the scheduled run and keeps its
history. Prompts are paused rather than deleted, so a prompt's history is never
lost.
Plans cap prompts per Website. An add over the limit is refused with the reason, never silently trimmed.
run_prompt_check is the spend tool for a single prompt. It checks one prompt
now across every assistant enabled for the Website, priced per assistant
checked. An assistant whose check fails is refunded and reported by name only,
without the upstream error text.
Finding prompts worth tracking
research_prompts discovers the questions people ask AI assistants in your
market, with an estimated demand figure and a relevance score for the Website.
It works from seed topics or from the brand profile, and it spends only on a
fresh run: repeating a recent search is served from cache at no cost.
The demand figure is modelled from Google's People Also Ask results. It is not a count of ChatGPT or Gemini queries, and the response says so.
list_prompt_candidates reads what an earlier run found, best first, for free.
Call it before paying for fresh research, and turn a candidate into a tracked
prompt with create_prompt.
AI Shopping
The AI Shopping tools measure products rather than the brand: whether your
monitored products appear when a shopper asks an AI assistant for a
recommendation. Shopper questions are prompts too, created with create_prompt
using the shopping kind and listed with list_prompts.
get_shopping_visibility reports, per surface over the last 28 days, how often
your products were recommended. It divides by answers that showed products at
all, because an answer with no product carousel says nothing about your
catalogue. A null means no answer showed products in the window, and it is
never rendered as zero.
list_shopping_competitors lists the merchants recommended instead of you. It
is built from product cards that did not match your catalogue, so an
incomplete catalogue inflates the list; say so when reporting it.
get_product_shopping_detail shows one product: which shopper questions it
appears on, which merchants shared those answers, and the cards the surface
showed. get_shopping_prompt_detail shows one shopper question: how often its
answers showed a carousel, whether your products were in it and at what
position, and who took the slots instead. Every share divides by that question's
own carousels.
run_shopping_check checks one shopper question now, outside its schedule,
priced per enabled shopping surface. It runs in the background, so read the
result afterwards with get_shopping_prompt_detail. One question can be checked
by hand once every five minutes.
Example prompts
Which competitors get named in answers where my brand does not? Use my saved competitor set.
Show me my tracked prompts and flag any that have never produced a known verdict.
Research the questions people ask AI assistants about running shoes, then show me the ten most relevant candidates before tracking any.
Pause the five prompts with the lowest mention rate.
Are my products recommended in AI shopping answers? If not, which merchants take the slots, and could an incomplete catalogue explain it?
Where to go next
- AI visibility tracking and AI Shopping visibility, for what RankX AI measures and how to start.
- Rankings and Google data, for the search side.
- Metrics defined, for what each figure means.
- Null is not zero, for the rule these tools all obey.
Questions
Does adding a tracked prompt cost credits?
Creating the prompt is free, but every active prompt is checked on a schedule and each check is charged. Pause a prompt with set_prompt_status to stop that cost while keeping its history.
Why is my mention rate null instead of 0%?
RankX AI divides by checks with a known verdict. A null rate means no check in the window produced one, so the result is unknown, which is different from never being named. See null is not zero.
Which AI assistants do these tools report on?
The assistants enabled for that Website: ChatGPT by default, plus any of Gemini, Claude and Perplexity switched on. Google AI Overviews are measured separately by the rank tracker, in Rankings and Google data.
Last updated