docs(features): show PostgreSQL notices as server output

databases/postgresql.mdx

@@ -175,6 +175,10 @@ A tab bound to a database that has no entry runs on a [metadata connection](/con
 
 In `pg_stat_activity`, each database's own connection reports `application_name` as `TablePro`, one per database entry in the strip. The [metadata connections](/connections/connection-form#connection-health) that read the sidebar and the Structure tab, and the ones a tab on a database without an entry runs on, report `TablePro Metadata`. To name them yourself, put `-c application_name=reporting` in **Connection Options**: every connection then uses that name.
 
+## Notices
+
+`RAISE NOTICE`, `RAISE WARNING` and the other notices the server sends show with the result of the statement that raised them, as psql prints them. See [Server output](/features/query-results#server-output).
+
 ## Text encoding
 
 Sessions run in `UTF8` whatever encoding the database was created with, and the server converts text both ways. Japanese in an `EUC_JP` database or accented names in a `LATIN1` one read and save intact, and stay that way after `RESET ALL` or `DISCARD ALL` in a query tab. A `-c client_encoding=…` in **Connection Options** is overridden; other settings there still apply.

features/explain-visualization.mdx

@@ -50,7 +50,9 @@ Then compare estimate against measurement. A node whose **Rows** estimate is ord
 
 ## View modes
 
-The segmented control above the results switches between three views. A plan the parser cannot read as a tree has no control at all and shows the raw output under a banner saying so.
+The segmented control above the results switches between three views. A plan the parser cannot read as a tree shows the raw output under a banner saying so, and has no control unless it also has output.
+
+`EXPLAIN ANALYZE` and an actual plan run the statement, so a plan whose statement printed on the server, such as a function's `RAISE NOTICE`, also offers **Output**. See [Server output](/features/query-results#server-output).
 
 The mode and the Compare baseline stay with the editor tab, so another plan in it opens in the same mode. Each plan keeps its own selected step, zoom and scroll position while its result is open. Running the statement again replaces an unpinned plan with a new one at 100%.
 

features/query-results.mdx

@@ -98,13 +98,21 @@ INSERT, UPDATE, DELETE and DDL show a success view with the affected row count a
 
 ## Server output
 
-Lines a statement prints on the server show with that statement's result. On Oracle that is `DBMS_OUTPUT.PUT_LINE`, and the output is on from the moment you connect: no `SET SERVEROUTPUT ON`.
+Lines a statement prints on the server show with that statement's result.
+
+| Engine | What shows | Kept per statement |
+| --- | --- | --- |
+| Oracle | `DBMS_OUTPUT.PUT_LINE` lines. Output is on from the moment you connect: no `SET SERVEROUTPUT ON` | 10,000 lines |
+| PostgreSQL | Notices such as `RAISE NOTICE`, `RAISE WARNING` or `VACUUM VERBOSE` lines, as psql prints them: `NOTICE:  step 1 done`, with their `DETAIL` and `HINT` lines. `client_min_messages` decides which levels the server sends | 10,000 lines |
+| SQL Server | `PRINT` output and messages up to severity 10 | 1,000 messages |
 
 - A block or procedure call that returns no rows shows its lines under the success message.
 - A query that returned rows and also printed gains an **Output** mode beside **Data**.
+- A plan from `EXPLAIN ANALYZE` or an actual plan gains **Output** beside its views, since the statement ran. See [EXPLAIN Visualization](/features/explain-visualization#view-modes).
+- **Fetch All** runs the query again and shows what that run printed.
 - A statement that fails lists the first 20 lines it printed before the error, inside the error banner.
 
-The lines are read after every statement run from the editor, so each result holds only what its own statement printed. At most 10,000 lines are kept per statement; past that the result reads "The output was cut short." and the rest is discarded. Lines a trigger prints while a grid edit is saved are not read then, and show with the next statement you run from the editor.
+Each result holds only what its own statement printed. When a script runs inside a transaction the app opens, what its closing commit prints, such as a deferred trigger's notices, shows with the last statement's result. Lines printed by anything else on the connection are not shown: a trigger fired by saving a grid edit, **Commit** in the toolbar, or a statement an AI assistant or MCP client ran. Past the limit, the result reads "The output was cut short." and the rest is discarded.
 
 ## Failed statements