> ## Documentation Index > Fetch the complete documentation index at: https://docs.fullreach.ai/llms.txt > Use this file to discover all available pages before exploring further. # Prompts > The questions this project asks the AI platforms every day, how each one does, and the questions the product suggests adding. A prompt is one question in one market and one language. It runs every day on every platform of the project. The plan counts prompts, pooled across the organization's projects. See [Markets and languages](/reference/markets-and-languages) and [Plan limits](/reference/limits). ## The answer at the top The page asks whether the answers name this brand when people ask these questions. It answers yes when at least one answer in the period names this brand. It also says how many prompts get no answer that names it. The line under it counts the running prompts. It also counts the prompts whose answers name no monitored brand in the period, and the paused prompts. ## The columns The table reads the period that the filter picks. Each figure counts that prompt's analysed answers. | Column | What it counts | | -------------- | ---------------------------------------------------------------------------------------- | | Visibility | Answers that name this brand, out of the prompt's answers, shown as "mentions / answers" | | Share of voice | This brand's mentions, out of every monitored brand's mentions in the prompt's answers | | Sentiment | The positive share of this brand's labelled mentions | | Recommendation | The recommended share of this brand's labelled mentions | | Names a brand | Answers that name this brand or any monitored competitor, out of the prompt's answers | | Own site | Answers that cite this brand's own website, out of the prompt's answers | | Last answer | The newest day with an answer | "Names a brand" tells two low visibilities apart. When few answers name any brand, the question is one that nobody wins yet. When most answers name a brand but not this one, the rivals win it. The competitor filter does not change this page. Share of voice here always counts every monitored brand. ## The marks under a prompt | Mark | Meaning | | -------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | | Topic, persona | The prompt's [topic](/reference/topics) and [persona](/reference/personas) | | paused | The prompt is not asked any more. What it measured stays on the row | | branded | The wording names this brand. Every figure leaves it out unless the branded filter asks for it. See [Branded prompts](/reference/filters#branded-prompts) | | names nobody | No answer named a company of any kind on the prompt's last 7 collection days | **names nobody** is stricter than "Names a brand". It counts every company an answer names, monitored or not, and it reads the prompt's own last collection days, whatever the filters say. A prompt that names nobody measures nothing about visibility. Reword it towards a question that asks for providers, or replace it with a suggestion below. The sidebar badge on Prompts counts the running prompts that name nobody. The day a prompt starts to name nobody, an Alert mail says so. ## Adding and changing prompts * **Add prompt** takes the wording, the market, the language, and optionally a topic and a persona. A prompt that already runs in that market with the same words and language is refused. * **Edit** changes the wording, the language, the topic or the persona. The market cannot change, because the history belongs to it. History stays on the prompt, under the wording that produced it. A new wording, or a new persona sentence, restarts the prompt's trend. * **Copy to market** copies prompts into one more market. See [Markets and languages](/reference/markets-and-languages#adding-and-pausing-a-market). * **View sources** opens the Sources page, filtered to this prompt. * **Pause** stops asking the prompt and frees its place in the plan. **Resume**, the ▶ on a paused row, asks it again from the next collection, when the plan has room. * **Delete** works only for a prompt that never ran. A prompt with history can only be paused. A row opens its latest answers, failures and "No AI Overview" included, with a link to the full history. ## Prompt History **See full history** lists every answer to one prompt, newest first, 50 at a time. Only the period, platform and market filters apply. The count includes failures and answers where Google showed no AI Overview, so it can be higher than the answers that other pages count. Each answer shows its platform, market, date and result: **Mentioned**, **Absent**, **Technical failure** or **No AI Overview**. A failure and "No AI Overview" count in no figure. **View answer** opens [the answer](/reference/answer-view). ## Suggestions **Questions worth adding** proposes new prompts. Each suggestion comes from one thing that this project's own answers showed, and it shows that thing as its evidence. | Reason | What the answers showed | | ------------- | ---------------------------------------------------------------------------------- | | Gone quiet | A prompt that names nobody. The suggestion replaces it | | Rivals cited | A page that answers cite beside rivals, and never beside this brand | | Search | A web search that an assistant ran, whose wording names a rival and not this brand | | Product cards | A rival's product on shopping answers that showed no product of this brand | * **Suggest** reads the answers of the last 7 days, without branded prompts, and writes the questions. It takes up to 3 signals of each reason per press. [Plan limits](/reference/limits) states how many presses an organization has per day. The panel says when they are spent. * **Track this** adds the suggestion as a prompt. A suggestion that replaces a quiet prompt also offers **Track this and pause the old one**, which frees the old prompt's place in the plan. * **×** dismisses a suggestion for good. After 3 dismissals, a signal gets no new questions until its evidence changes. A suggestion carries a topic and a persona. A replacement keeps those of the prompt it replaces. A suggestion never creates a topic. Before you accept anything, the panel says how the batch splits between the personas. ## Where it appears | Surface | Form | | -------------------------------------------------------------- | ----------------------------------------------------------------------------------- | | The Prompts page | The table, the marks, the suggestions and each prompt's history | | The weekly Digest | Prompts that started or stopped naming this brand, and the prompts that name nobody | | An Alert mail | The day a prompt starts to name nobody | | MCP, `list_prompts`, `list_prompt_runs` and `list_suggestions` | The same rows, the answers to one prompt and the suggestions | | CSV export | One line per prompt |