Prompts

What a prompt is

A prompt is a full question, stored as its own row, that DiscoveredBy actually runs against the AI engines you track, not a fragment of one. A keyword is a different thing: a plain text label belonging to the project, which you can attach to a prompt to file it, with no country, no schedule, and nothing that runs it. See Key terms for how a prompt fits the rest of the product's vocabulary.

One target per country

Every prompt tracks at least one country. Each country you add creates its own row underneath the prompt, called a prompt target, and that row, not the prompt itself, is what actually gets run: a prompt tracked in three countries produces three prompt targets, each checked independently against every engine on your plan. Adding a country to a prompt, or reactivating one you had paused, uses one of your plan's prompt slots; see Pricing for what each plan includes.

How prompts get run

Running a prompt is not something you trigger yourself, with one exception: accepting a prompt from Suggestions immediately queues a fast first run outside the daily cycle (see Your first scan for what that exception does and does not change). Otherwise, a single scheduled job, once a day, enqueues one run per active prompt target, on every engine currently available on your plan. It requires the project, the prompt and the target all to be active before it will enqueue anything, so pausing a prompt, or turning off one of its countries, removes it from that job's next run. See Your first scan for how long the first run takes and what appears once it finishes.

The prompt list

Prompts lists everything you are tracking: each row carries its own score, a trend sparkline, citation rate, the countries and engines it covers, its tags and its keywords as chips, and a running or paused toggle you can flip without leaving the list. The heading above the table averages that score across every tracked prompt, and counts how many are currently running versus paused.

Four dropdowns sit above the table, and they are the whole set: status, tag, keyword and country. Each one narrows what is left rather than widening it, so a prompt has to satisfy every filter you have set at once (frontend/src/routes/(app)/prompts/+page.svelte#keywordFilter). The keyword dropdown and the chips on each row both exist because the list endpoint returns a prompt's keywords alongside its tags (api/routers/prompts.py#list_prompts, api/schemas/prompt.py#keywords).

Not every prompt on the list has to start with you typing one in. Prompts → Suggestions is a queue of proposed prompts you track or dismiss, badge-counted on this same screen; that page covers where they come from.

Inside one prompt

Opening one prompt leads with its citation rate: how many of that prompt's completed runs in the last 30 days produced an answer citing your domain, out of how many completed in that same window (api/routers/pages.py#prompt_detail_page). That is citation rate exactly as defined elsewhere in the product, scoped down to one prompt instead of the whole project, and the screen now names it that. The same figure used to appear here as a "visibility" score out of 100, from a second field carrying this rate multiplied by 100 under a name that belongs to a different metric; that field was removed rather than renamed, so one number is left under one name (api/schemas/prompt_detail.py#citation_rate). With no completed run inside that window there is nothing to show, not a zero; see How to read any number here.

Two engine callouts sit beside it, the strongest and the weakest engine for this prompt specifically. Both cards always show; with fewer than two engines to compare, the weakest one has nothing to report and says so instead of showing a second score.

Below that: which pages on your own site you have mapped to this prompt, an engine-by-engine breakdown of the same score, a tally of every brand an engine has cited when answering this question (yours included), and the individual citations behind that tally, grouped by engine. How an answer expanded into an engine's own sub-queries, where the engine reports it, gets its own panel here too; see Fan-out for what that panel shows, including how a captured query gets classified and when one does not, and Engines and measurement for what the expansion itself means. A run history table closes out the page, capped at the 50 most recent runs with a note when there are more, because the underlying record itself carries no such limit.

Related

• Suggestions and opportunities: where suggested prompts come from, and what a Search Console opportunity is
• Fan-out: what each engine actually searched for behind this prompt, and the eight categories those searches fall into
• Key terms: what counts as a prompt, a mention, and a citation
• Your first scan: the daily run in full, and how long the first one takes
• Engines and measurement: which engines run, and what fan-out means
• Metrics defined: every project-wide number's exact formula and denominator