feat(briefing): wire pre-processing pipeline; run_compilation returns (text, metadata)

- Task change detection via snapshot diff
- RSS scoring/filtering via briefing_preferences
- Weather card via parse_weather_card_data (staleness-gated)
- News card markdown format with ordering constraint
- Metadata stored on Message record via post_message()

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/fabledassistant/routes/briefing.py

@@ -232,6 +232,6 @@ async def manual_trigger():
 
     model = await get_setting(g.user.id, "default_model", "")
     conv = await get_or_create_today_conversation(g.user.id, model)
-    text = await run_compilation(g.user.id, slot, model)
+    text, metadata = await run_compilation(g.user.id, slot, model)
-    msg = await post_message(conv.id, "assistant", text)
+    msg = await post_message(conv.id, "assistant", text, metadata=metadata)
     return jsonify({"conversation_id": conv.id, "message_id": msg.id, "slot": slot})

src/fabledassistant/services/briefing_pipeline.py

@@ -248,24 +248,36 @@ def _internal_system_prompt(profile_body: str) -> str:
 
 def _external_system_prompt() -> str:
     return (
-        "You are a briefing assistant for external information. Your job is to summarise "
+        "You are a briefing assistant for external information. Your job is to present "
-        "the user's RSS feed digest and weather forecast into a concise, engaging update. "
+        "selected news items and summarise any remaining RSS content. "
-        "Group related news items. Note any significant weather changes. "
+        "IMPORTANT: Weather is handled separately — do NOT include any weather section.\n\n"
-        "Be informative but brief. Do not discuss tasks, calendar, or work items."
+        "Format each news item EXACTLY as:\n"
+        "**[Headline text](source_url)**\n"
+        "*Outlet Name · Day Month*\n"
+        "One or two sentence summary.\n\n"
+        "Present news items in the EXACT ORDER they are provided. Do not reorder them. "
+        "After the news cards, add a brief paragraph for any remaining context."
     )
 
 
 def _internal_user_prompt(data: dict, slot: str) -> str:
     lines = [f"Briefing slot: {slot}", f"Date: {data['date']}", ""]
-    if data["overdue_tasks"]:
+    if data.get("unchanged_task_count", 0) > 0:
-        lines.append(f"OVERDUE ({len(data['overdue_tasks'])}):")
+        lines.append(
-        lines.extend(f"  - {t}" for t in data["overdue_tasks"])
+            f"({data['unchanged_task_count']} tasks are unchanged since the last briefing "
+            "— acknowledge briefly, do not list them.)"
+        )
         lines.append("")
-    if data["due_today"]:
+    changed = data.get("changed_tasks") or data.get("overdue_tasks", [])
+    if changed:
+        lines.append(f"CHANGED/NEW TASKS ({len(changed)}):")
+        lines.extend(f"  - {t}" for t in changed)
+        lines.append("")
+    if data.get("due_today"):
         lines.append(f"DUE TODAY ({len(data['due_today'])}):")
         lines.extend(f"  - {t}" for t in data["due_today"])
         lines.append("")
-    if data["high_priority"]:
+    if data.get("high_priority"):
         lines.append("HIGH PRIORITY (in progress):")
         lines.extend(f"  - {t}" for t in data["high_priority"])
         lines.append("")
@@ -326,53 +338,104 @@ async def _get_temp_unit(user_id: int) -> str:
         return "C"
 
 
-async def run_compilation(user_id: int, slot: str, model: str | None = None) -> str:
+async def run_compilation(
+    user_id: int,
+    slot: str,
+    model: str | None = None,
+) -> tuple[str, dict]:
     """
     Run the full two-lane briefing pipeline for a user and slot.
-    Returns the combined briefing text to be posted as the opening assistant message.
+    Returns (briefing_text, metadata_dict) where metadata contains
+    weather card data and rss_item_ids for frontend rendering.
     """
     if model is None:
         model = await get_setting(user_id, "default_model", Config.OLLAMA_MODEL)
 
     from fabledassistant.services.briefing_profile import get_profile_body
+    from fabledassistant.services.briefing_preferences import (
+        load_topic_preferences,
+        load_topic_reaction_scores,
+        score_and_filter_items,
+    )
+    from fabledassistant.services.weather import parse_weather_card_data, get_cached_weather_rows
+
     profile_body, temp_unit = await asyncio.gather(
         get_profile_body(user_id),
         _get_temp_unit(user_id),
     )
 
-    # Parallel gather
+    # ── Pre-processing ──────────────────────────────────────────────────────────
-    internal_data, external_data = await asyncio.gather(
+    include_topics, exclude_topics = await load_topic_preferences(user_id)
+    topic_scores = await load_topic_reaction_scores(user_id)
+
+    # Parallel raw gather — weather rows fetched in same gather to avoid extra DB round-trip
+    internal_data, external_data, weather_rows = await asyncio.gather(
         _gather_internal(user_id),
         _gather_external(user_id),
+        get_cached_weather_rows(user_id),
     )
 
-    # Two-lane LLM synthesis (both calls run concurrently)
+    # Task change detection
+    all_tasks = internal_data.get("all_tasks_raw", [])
+    changed_tasks, unchanged_count = await split_changed_tasks(user_id, all_tasks)
+
+    # RSS filtering
+    raw_rss = external_data.get("rss_items") or []
+    filtered_rss = score_and_filter_items(
+        raw_rss,
+        include_topics=include_topics,
+        exclude_topics=exclude_topics,
+        topic_scores=topic_scores,
+        max_items=10,
+    )
+    rss_item_ids = [item["id"] for item in filtered_rss if item.get("id")]
+
+    # Weather staleness gate — returns None if data is >24h old
+    weather_card = parse_weather_card_data(weather_rows[0], temp_unit) if weather_rows else None
+
+    # ── LLM Synthesis ──────────────────────────────────────────────────────────
+    # Build filtered internal data with only changed tasks
+    today = internal_data["date"]
+    internal_data_filtered = dict(internal_data)
+    internal_data_filtered["unchanged_task_count"] = unchanged_count
+    internal_data_filtered["changed_tasks"] = [format_task(t) for t in changed_tasks]
+
+    # Build filtered external data (suppress weather prose — card handles it)
+    external_data_filtered = {
+        "rss_items": filtered_rss,
+        "weather": [],
+    }
+
     internal_text, external_text = await asyncio.gather(
         _llm_synthesise(
             _internal_system_prompt(profile_body),
-            _internal_user_prompt(internal_data, slot),
+            _internal_user_prompt(internal_data_filtered, slot),
             model,
         ),
         _llm_synthesise(
             _external_system_prompt(),
-            _external_user_prompt(external_data, slot, temp_unit),
+            _external_user_prompt(external_data_filtered, slot, temp_unit),
             model,
         ),
     )
 
+    # ── Post-processing ─────────────────────────────────────────────────────────
+    await upsert_task_snapshots(user_id, all_tasks)
+
+    metadata: dict = {"rss_item_ids": rss_item_ids, "weather": weather_card}
+
     if not internal_text and not external_text:
         logger.warning("Briefing compilation produced no content for user %d slot %s", user_id, slot)
-        return ""
+        return "", metadata
 
     greeting = slot_greeting(slot)
-    today = internal_data["date"]
     parts = [f"**{greeting} — {today}**", ""]
     if internal_text:
         parts += ["## Your Day", "", internal_text, ""]
     if external_text:
         parts += ["## The World", "", external_text]
 
-    return "\n".join(parts).strip()
+    return "\n".join(parts).strip(), metadata
 
 
 async def run_slot_injection(user_id: int, slot: str, model: str | None = None) -> str:

src/fabledassistant/services/briefing_scheduler.py

@@ -153,9 +153,9 @@ async def _run_slot_for_user(user_id: int, slot: str) -> None:
 
         await _run_profile_closeout(user_id, model)
         conv = await get_or_create_today_conversation(user_id, model)
-        text = await run_compilation(user_id, slot, model)
+        text, metadata = await run_compilation(user_id, slot, model)
         if text:
-            await post_message(conv.id, "assistant", text)
+            await post_message(conv.id, "assistant", text, metadata=metadata)
 
     else:
         conv = await get_or_create_today_conversation(user_id, model)