AI-Ready Databases with Semantic Intelligence

Analyse your data with your AI client · AI-ready Semantic Intelligence

The Semantic Intelligence you ran makes every AI client sharper

Navigation by business area, profiles without a scan, column-by-column control over trimming, and live profiles that keep the analysis current.

Personal Core and up Runs on your Network Agent Tylko do odczytu, zawsze

Utknąłeś w jakimś momencie? Po prostu zapytaj nas. Otwórz Wsparcie w Query Streams (menu w prawym górnym rogu → Wsparcie) i wyślij nam wiadomość. Przeczytają ją prawdziwi ludzie, a pytanie „Nie wiem, o co pyta ten ekran” jest jak najbardziej trafne.

What your Network Agent does for you

A database where Semantic Intelligence has run is AI-ready: your AI client navigates it by business area, gets profiles of plain tables straight from the analysis, and gets results trimmed in the right places. The groundwork Semantic Intelligence did once is reused on every question.

Co otrzymujesz	How Semantic Intelligence helps
Navigation by business area	The business areas your AI client sees first come from the Semantic Intelligence analysis. See Schema that fits.
Profiles without a scan	A profile of one whole table can be answered from the analysis: most common values, distributions and samples. See Profile a result.
Trimming in the right places	Each column’s AI results menu on the Semantic Intelligence page can keep it always in full, never trim it, trim it like the rest, or hide it as binary.
Fewer wrong first queries	Descriptions, value lists and discovered joins mean your AI client picks the right column the first time.
Up-to-date analysis	A live profile on a plain table refreshes the stored statistics for those columns, and the Semantic Intelligence page notes when an AI client’s profile last refreshed a column.

The plan it needs

Plan	This capability	Uwagi
Bezpłatny	Not on this plan	Works as it always has, with standard results.
Personal Core, Personal Plus	Dołączony	On every database where Semantic Intelligence has run.
Business Core, Business Plus, Business Max	Dołączony	Part of Context management and Result analysis.
Enterprise Core, Plus, Max	Dołączony	Part of Context management and Result analysis.

Why it is on paid plans: your Network Agent does the heavy lifting next to your database, so the rows stay in your network and your AI client spends its attention on the thinking. That groundwork is what the paid plans add.

Semantic Intelligence itself is run from Nova. What this page describes uses it on plans with Context management and Result analysis.

The Free plan works as it always has. Paid plans add the Analytic toolset. Free is a plan you keep: MCP, the REST API, federated queries and alerts, with the same results as before. Plan names and prices are on the strona z cenami.

How to ask

You do not call tools yourself. Ask in plain words and your AI client picks the tool. Three asks that bring this capability in:

Klaudiusz

ChatGPT

Kursor

Kod Claude'a+10 moreTylko do odczytu

Profile the carrier table.

qs_run_query · mode=profile

Use today’s data for that, not the stored analysis.

qs_run_query · mode=profile · prefer_live

Is the sales database ready for this?

qs_get_si_status

Groundwork done on your Network Agent, inside your network zalogowano na karcie Aktywność MCP

For people who read tool lists: no new tool. A profile answered from the analysis says profile_source: semantic_intelligence; prefer_live forces a live read. qs_get_si_status reports coverage per database, and tabela_profili_qs reads a table’s stored profile on every plan. See qs_profile_table on the Tools tab.

What comes back

Klaudiusz

ChatGPT

Kursor

Kod Claude'a+10 moreTylko do odczytu

Profile the carrier table.

qs_run_query · mode=profile inteligencja_schematu

From your Semantic Intelligence analysis, no scan needed:

column	values
carrier_code	CARRIER-A 44% · CARRIER-B 31% · CARRIER-C 25%
service_level	standard 80% · express 20%
aktywny	yes 96% · no 4%

The analysis is three months old. Want today’s figures instead? That reads the table live.

Groundwork done on your Network Agent, inside your network answered from the analysis · no rows read

profile_source: semantic_intelligence for a stored profile, live for a read.
Coverage: When the analysis ran and how many columns it has statistics for, so your AI client can decide whether to read live.

What it does not help with

A stored profile is as old as the analysis. Ask for today’s data when it matters.
Only plain reads of one table qualify. Joins, filters and calculated columns are profiled live.
Columns your organization excluded from AI are not used.
It does not run Semantic Intelligence for you. Run it once per database that matters.

Ograniczenia

Navigation by business area needs Context management; stored profiles need Result analysis. Both are on Personal Core and up.
Tables whose analysis holds no value statistics are read live instead.
Relationship and statistics scans run after every schema capture on every plan; the full Semantic Intelligence analysis is run from Nova.

Ustawienia

Run or refresh Semantic Intelligence from Nova → Inteligencja, where an AI-ready badge marks databases whose profiles and compact schemas are served instantly to AI clients. Each column’s AI results menu sets how it is trimmed: Always keep full text, Never trim, Trim like the rest lub Hide as binary. Jak działa inteligencja semantyczna.

When your AI client says no

When the server refuses, your AI client receives a code and one sentence, and usually relays it. Quoted sentences are the server’s own words; the parts in italics stand for the ids a real message names.

Kod	What the server says	Co robić
`MCP_FEATURE_NOT_INCLUDED`	“Your plan does not include result analysis; profiles, data-quality reports and follow-up questions on a held result are available on Personal Core.”	Ask for rows instead, or move up to Personal Core.
`BAZA_DANYCH_MCP_NIE_PRZECHWYCONA`	The database is outside the connector’s capture selection, so there is no analysis to use.	Add it to the capture selection, then run Semantic Intelligence.

Przykłady

“What statuses can an order have?”

The answer comes from the analysis, with each status and its share, and the first query that filters on it uses the right value.

“List customers with their notes.”

Notes are trimmed with a marker; the customer number column, set to Never trim on the Semantic Intelligence page, is never cut.

Nadal masz problem? Otwórz Wsparcie w Query Streams i powiedz nam, na jakim etapie jesteś i co widzisz. Połączymy Cię.

Facts source

Setup trouble?

Agent shows Offline, or Test connection and Browse are greyed out? Check that the Agent is online → Is the Agent online?