fix: scope thread detection transcript to closing scene (T80.2)

apply_scene_close_summary fed detect_threads the chat-wide last-50
turns. When a chat has accumulated multiple scenes' worth of dialogue,
that bleeds prior-scene turns into the second close's classifier prompt
and risks mis-attributing threads (closing one that opened earlier,
re-opening one that already closed).

Add an optional ``since_event_id`` kwarg to ``_read_recent_dialogue``
that lower-bounds by event_log id, plus a ``_scene_opened_event_id``
helper that resolves the scene-open event for a given scene_id. Wire
both into the thread-detection call site so its scene_transcript
holds only the closing scene's turns. The per-POV summarizer keeps the
chat-wide approximation it had before — that's intentional.

Adds test_thread_detection_uses_scene_scoped_transcript.

chat/services/scene_summarize.py

@@ -123,7 +123,11 @@ async def summarize_scene(
 
 
 def _read_recent_dialogue(
-    conn: Connection, chat_id: str, *, limit: int = 50
+    conn: Connection,
+    chat_id: str,
+    *,
+    limit: int = 50,
+    since_event_id: int | None = None,
 ) -> list[dict]:
     """Pull the last ``limit`` user/assistant turns for ``chat_id``.
 
@@ -132,14 +136,29 @@ def _read_recent_dialogue(
     the most recent turns of the chat. Superseded and hidden rows are
     filtered out so regenerated turns (T29) don't bleed into the
     summary.
+
+    T80.2: ``since_event_id`` clamps the result to event_log rows whose
+    ``id >= since_event_id`` so callers needing a scene-scoped view (e.g.
+    thread detection on close) don't pull turns that landed before the
+    closing scene's ``scene_opened`` event.
     """
-    cur = conn.execute(
+    if since_event_id is None:
-        "SELECT kind, payload_json FROM event_log "
+        cur = conn.execute(
-        "WHERE kind IN ('user_turn', 'assistant_turn') "
+            "SELECT kind, payload_json FROM event_log "
-        "  AND superseded_by IS NULL AND hidden = 0 "
+            "WHERE kind IN ('user_turn', 'assistant_turn') "
-        "ORDER BY id DESC LIMIT ?",
+            "  AND superseded_by IS NULL AND hidden = 0 "
-        (limit,),
+            "ORDER BY id DESC LIMIT ?",
-    )
+            (limit,),
+        )
+    else:
+        cur = conn.execute(
+            "SELECT kind, payload_json FROM event_log "
+            "WHERE kind IN ('user_turn', 'assistant_turn') "
+            "  AND superseded_by IS NULL AND hidden = 0 "
+            "  AND id >= ? "
+            "ORDER BY id DESC LIMIT ?",
+            (since_event_id, limit),
+        )
     rows = list(reversed(cur.fetchall()))
     out: list[dict] = []
     for kind, payload_json in rows:
@@ -158,6 +177,65 @@ def _read_recent_dialogue(
     return out
 
 
+def _scene_opened_event_id(
+    conn: Connection, chat_id: str, scene_id: int
+) -> int | None:
+    """Return the event_log id of the ``scene_opened`` (or
+    ``meanwhile_scene_started``) event that created scene row
+    ``scene_id``. Used by T80.2 to lower-bound dialogue reads to a
+    single scene's transcript.
+
+    ``meanwhile_scene_started`` carries an explicit ``scene_id`` so we
+    match on that directly. ``scene_opened`` doesn't, so we walk the
+    chat's scene rows in id order and zip against the chat's scene-open
+    events in id order — the projector creates one scene row per
+    scene-open event, so positions correspond.
+
+    Returns ``None`` when no matching event is found; callers should
+    treat that as "fall back to chat-wide" rather than over-filter.
+    """
+    # Fast path for meanwhile children (explicit scene_id in payload).
+    for ev_id, payload_json in conn.execute(
+        "SELECT id, payload_json FROM event_log "
+        "WHERE kind = 'meanwhile_scene_started' "
+        "  AND superseded_by IS NULL AND hidden = 0",
+    ).fetchall():
+        try:
+            p = json.loads(payload_json)
+        except (TypeError, ValueError):
+            continue
+        if p.get("chat_id") == chat_id and p.get("scene_id") == scene_id:
+            return ev_id
+    # Fallback for parent you-scenes: zip chat-scoped scene-open events
+    # against chat-scoped scene rows in id order.
+    chat_scene_ids = [
+        r[0]
+        for r in conn.execute(
+            "SELECT id FROM scenes WHERE chat_id = ? ORDER BY id ASC",
+            (chat_id,),
+        ).fetchall()
+    ]
+    if scene_id not in chat_scene_ids:
+        return None
+    chat_open_evs: list[int] = []
+    for ev_id, _kind, payload_json in conn.execute(
+        "SELECT id, kind, payload_json FROM event_log "
+        "WHERE kind IN ('scene_opened', 'meanwhile_scene_started') "
+        "  AND superseded_by IS NULL AND hidden = 0 "
+        "ORDER BY id ASC",
+    ).fetchall():
+        try:
+            p = json.loads(payload_json)
+        except (TypeError, ValueError):
+            continue
+        if p.get("chat_id") == chat_id:
+            chat_open_evs.append(ev_id)
+    idx = chat_scene_ids.index(scene_id)
+    if idx < len(chat_open_evs):
+        return chat_open_evs[idx]
+    return None
+
+
 async def _summarize_and_apply_for_witness(
     conn: Connection,
     client: LLMClient,
@@ -487,16 +565,27 @@ async def apply_scene_close_summary(
                 },
             )
 
-    # T58.2: thread detection on close. Reuses the dialogue we already
+    # T58.2: thread detection on close. Failure-tolerant: classify()
-    # gathered for per-POV summarization — same {speaker, text} shape
+    # returns the empty default on retry-exhaustion, and the broad except
-    # detect_threads expects. Failure-tolerant: classify() returns the
+    # below protects the close pipeline from any other classifier/mock
-    # empty default on retry-exhaustion, and the broad except below
+    # flap.
-    # protects the close pipeline from any other classifier/mock flap.
+    #
+    # T80.2: thread detection runs against a SCENE-SCOPED transcript,
+    # not the chat-wide last-50 turns used by the per-POV summaries.
+    # Mis-attributing threads when scene boundaries fall inside the last
+    # 50 turns would otherwise close threads opened in a prior scene.
+    scene_open_ev_id = _scene_opened_event_id(conn, chat_id, scene_id)
+    if scene_open_ev_id is not None:
+        scene_dialogue = _read_recent_dialogue(
+            conn, chat_id, since_event_id=scene_open_ev_id
+        )
+    else:
+        scene_dialogue = dialogue
     try:
         thread_result = await detect_threads(
             client,
             classifier_model=classifier_model,
-            scene_transcript=dialogue,
+            scene_transcript=scene_dialogue,
             open_threads=list_open_threads(conn, chat_id),
             timeout_s=timeout_s,
         )

tests/test_per_pov_summary.py

@@ -1490,3 +1490,138 @@ async def test_scene_close_re_run_does_not_double_suffix(tmp_path):
             # from a row whose text already contained the suffix).
             inner_count = pov.count("Key quotes:")
             assert inner_count == 1
+
+
+@pytest.mark.asyncio
+async def test_thread_detection_uses_scene_scoped_transcript(
+    tmp_path, monkeypatch
+):
+    """T80.2: when a chat has multiple closed scenes, the second scene's
+    close must hand ``detect_threads`` ONLY the second scene's turns —
+    not the chat-wide last-50, which would bleed in the first scene's
+    transcript and risk mis-closing threads."""
+    from chat.services import thread_detection as td_mod
+
+    canned = json.dumps(
+        {
+            "summary": "BotA had a quick chat.",
+            "knowledge_facts": [],
+            "relationship_summary": "Steady.",
+        }
+    )
+
+    captured_transcripts: list[list[dict]] = []
+
+    async def capturing_detect_threads(client, **kwargs):
+        captured_transcripts.append(list(kwargs["scene_transcript"]))
+        return td_mod.ThreadDetectionResult()
+
+    monkeypatch.setattr(td_mod, "detect_threads", capturing_detect_threads)
+
+    db = tmp_path / "t.db"
+    apply_migrations(db)
+    with open_db(db) as conn:
+        # Seed scene 1 + 3 turns + close.
+        _seed_single_bot_scene(conn)
+        # Add two extra distinct turns inside scene 1 so the transcript
+        # has clearly-scene-1 markers we can assert on.
+        append_event(
+            conn,
+            kind="user_turn",
+            payload={
+                "chat_id": "chat_bot_a",
+                "prose": "SCENE_ONE_USER_TURN",
+                "segments": [],
+            },
+        )
+        append_event(
+            conn,
+            kind="assistant_turn",
+            payload={
+                "chat_id": "chat_bot_a",
+                "speaker_id": "bot_a",
+                "text": "SCENE_ONE_BOT_TURN",
+                "truncated": False,
+                "user_turn_id": 2,
+            },
+        )
+        project(conn)
+
+        # Close scene 1.
+        client = MockLLMClient(canned=[canned])
+        await apply_scene_close_summary(
+            conn,
+            client,
+            classifier_model="x",
+            chat_id="chat_bot_a",
+            scene_id=1,
+            host_bot_id="bot_a",
+        )
+
+        # Open scene 2 with distinct dialogue. Use append_and_apply so
+        # the new events project incrementally without re-running the
+        # already-applied seed events.
+        from chat.eventlog.log import append_and_apply
+
+        append_and_apply(
+            conn,
+            kind="scene_opened",
+            payload={
+                "chat_id": "chat_bot_a",
+                "container_id": 1,
+                "started_at": "2026-04-26T21:00:00+00:00",
+                "participants": ["you", "bot_a"],
+            },
+        )
+        append_and_apply(
+            conn,
+            kind="memory_written",
+            payload={
+                "owner_id": "bot_a",
+                "chat_id": "chat_bot_a",
+                "scene_id": 2,
+                "pov_summary": "Original (scene 2)",
+                "witness_you": 1,
+                "witness_host": 1,
+                "witness_guest": 0,
+                "significance": 1,
+            },
+        )
+        append_and_apply(
+            conn,
+            kind="user_turn",
+            payload={
+                "chat_id": "chat_bot_a",
+                "prose": "SCENE_TWO_USER_TURN",
+                "segments": [],
+            },
+        )
+        append_and_apply(
+            conn,
+            kind="assistant_turn",
+            payload={
+                "chat_id": "chat_bot_a",
+                "speaker_id": "bot_a",
+                "text": "SCENE_TWO_BOT_TURN",
+                "truncated": False,
+                "user_turn_id": 3,
+            },
+        )
+
+        # Close scene 2.
+        client2 = MockLLMClient(canned=[canned])
+        await apply_scene_close_summary(
+            conn,
+            client2,
+            classifier_model="x",
+            chat_id="chat_bot_a",
+            scene_id=2,
+            host_bot_id="bot_a",
+        )
+
+        # The second close's transcript holds only scene-2 markers.
+        assert len(captured_transcripts) == 2
+        scene_two_transcript = captured_transcripts[1]
+        joined = " ".join(t.get("text", "") for t in scene_two_transcript)
+        assert "SCENE_TWO" in joined
+        assert "SCENE_ONE" not in joined