src/flask/static/css/app.css

@@ -716,6 +716,12 @@ table.data .comment { color: var(--text-muted); font-size: .84rem; }
     color: var(--amber-text);
 }
 
+/* The advice on the active-beers card and the run detail page - the
+   full text, not the clamped table cell. */
+.advice--reading {
+    margin-top: .5rem;
+}
+
 /* Advice that could not be generated - it says why. */
 .advice--failed {
     border-left-color: var(--danger);

src/flask/views/brews.py

@@ -60,7 +60,8 @@ def _start_advice(run_id: str, reading_id: str, corrected: bool = False) -> None
     comment is never touched.
 
     The background thread cannot use the request-scoped repository - that
-    client is closed when the request ends - so it builds its own.
+    client is closed when the request ends - so it uses the shared
+    process-wide one.
     """
     if not reading_id:
         return
@@ -130,7 +131,18 @@ def _start_advice(run_id: str, reading_id: str, corrected: bool = False) -> None
 
         try:
             shared = get_shared_repository()
-            shared.set_reading_advice(read_id, text, ok, expected_sg=advice_sg)
+            written = shared.set_reading_advice(read_id, text, ok,
+                                                expected_sg=advice_sg)
+            if not written:
+                # The reading was corrected while the advice was being
+                # generated, so the advice no longer matches it.  That is
+                # the guard working - the fresh advice for the corrected
+                # reading will land in its own turn.
+                log.info(
+                    "the advice for reading %s was not stored: the reading "
+                    "no longer holds the gravity it was written for",
+                    read_id,
+                )
         except Exception:  # noqa: BLE001 - never let the thread die noisily
             log.exception("could not store the generated advice")
 

tests/test_brew_advice.py

@@ -498,6 +498,20 @@ def test_the_advice_is_about_the_reading_that_was_saved():
         "the advice was generated about the wrong reading"
     )
 
+    # And the PROMPT must ask about that reading - not just pass it
+    # through.  Asserting only which dict was passed would not have
+    # caught the prompt still saying "give advice for the most recent
+    # reading", which is the bug that made the model advise about the
+    # wrong number.
+    prompt = brew_advice.build_prompt(
+        {"canonical_name": "Ale"},
+        captured["reading"],
+        captured["history"],
+    )
+    assert "THIS IS THE READING TO ADVISE ON" in prompt
+    assert "2026-01-01" in prompt
+    assert "most recent reading" not in prompt
+
 
 def test_the_shared_repository_is_built_only_once():
     """Concurrent advice threads must not each build their own client.