Opus Growth

Query Website Database

Runs a READ-ONLY SQL query against the website's own MariaDB database and returns the rows.

Type
Lecture seule
Crédits
1
Plateforme
AI Server
Validation
Inutile, rien ne change

Ce qu'il fait

Runs a READ-ONLY SQL query against the website's own MariaDB database and returns the rows. Only SELECT / SHOW / DESCRIBE / EXPLAIN / WITH are accepted; anything that changes data is refused here (use site_db_execute for that). `INTO OUTFILE` is refused too, because it writes files on the server. It inspects an application's data: how many orders/applications exist, what a table looks like, whether a migration landed. Databases only exist on application-type hosting servers; a static site has none. Every result row is returned: `limit` = rows in this response (no upper bound; 0 = as many as fit, about 370 KB); total_rows is the full count, and when has_more is true page_token=next_page_token returns the next rows without running the query again (offset=next_offset also works).

La description est en anglais car c'est le texte officiel que voit votre IA.

MCP: site_db_query

Comment l'utiliser

Vous n'appelez pas les outils par leur nom. Écrivez ce que vous voulez dans ChatGPT, Claude ou un autre assistant connecté à Opus Growth ; l'IA choisit cet outil quand il convient.

Paramètres

NomTypeObligatoirePar défaut
sqlstringFacultatif·Exact sql value for site db query. Supply text in the provider's required format; when it selects an existing resource, use the exact identifier returned by the relevant list or search operation.
serverstringFacultatif·Name of the connected hosting server to target when more than one server is available.
databasestringFacultatif·Exact database name or database selector on the managed site or server.
limitintegerFacultatif0Maximum number of rows returned in this response. When the result has_more flag is true, continue with next_page_token rather than assuming the list is complete.
offsetintegerFacultatif0Zero-based position at which to start reading results. Use the next_offset returned by the previous response when present.
page_tokenstringFacultatif·Continuation token returned as next_page_token by a previous call. Leave empty for the first page; pass it unchanged to retrieve the cached next page.