feat(briefing): cluster-level preference filtering — rank themes by user interest, suppress disliked topics

src/fabledassistant/services/briefing_pipeline.py

@@ -652,8 +652,9 @@ def _unified_user_prompt(internal_data: dict, external_data: dict, slot: str, te
         lines.extend(f"  - {p}" for p in internal_data["stale_projects"])
         lines.append("")
 
-    # News — clustered by topic for thematic synthesis
+    # News — clustered by topic, ranked by preference score
     rss = external_data.get("rss_items") or []
+    topic_scores = external_data.get("topic_scores") or {}
     if rss:
         # Group articles by their primary topic
         clusters: dict[str, list[dict]] = {}
@@ -666,19 +667,34 @@ def _unified_user_prompt(internal_data: dict, external_data: dict, slot: str, te
             else:
                 uncategorized.append(item)
 
-        # Sort clusters by size (most articles = most active theme)
-        sorted_clusters = sorted(clusters.items(), key=lambda x: len(x[1]), reverse=True)
+        # Score each cluster by aggregate preference (positive = user likes this topic)
+        def cluster_score(topic: str, items: list[dict]) -> float:
+            pref = topic_scores.get(topic, 0.0)
+            return pref * 2.0 + len(items)  # preference-weighted, size as tiebreak
+
+        # Filter out clusters where the user has a strongly negative preference
+        filtered_clusters = [
+            (topic, items) for topic, items in clusters.items()
+            if topic_scores.get(topic, 0.0) >= -2.0
+        ]
+
+        # Sort by preference-weighted score
+        sorted_clusters = sorted(filtered_clusters, key=lambda x: cluster_score(x[0], x[1]), reverse=True)
 
         lines.append("NEWS THEMES (synthesize 1-2 themes into your briefing; mention article count per theme):")
-        for topic, items in sorted_clusters[:4]:
+        for i, (topic, items) in enumerate(sorted_clusters[:4]):
             count = len(items)
-            titles = [i.get("title", "") for i in items[:3]]
-            sources = list({i.get("feed_title") or i.get("source") or "News" for i in items})
-            lines.append(f"  [{topic.title()}] ({count} article{'s' if count != 1 else ''} from {', '.join(sources[:2])})")
+            titles = [it.get("title", "") for it in items[:3]]
+            sources = list({it.get("feed_title") or it.get("source") or "News" for it in items})
+            pref_indicator = ""
+            pref = topic_scores.get(topic, 0.0)
+            if pref >= 2.0:
+                pref_indicator = " ★"  # user's preferred topic
+            lines.append(f"  [{topic.title()}]{pref_indicator} ({count} article{'s' if count != 1 else ''} from {', '.join(sources[:2])})")
             for t in titles:
                 lines.append(f"    • {t}")
-            # Include one excerpt for the top cluster
-            if items == sorted_clusters[0][1]:
+            # Include excerpt for preferred clusters or the top cluster
+            if pref >= 1.0 or i == 0:
                 excerpt = (items[0].get("content") or items[0].get("snippet") or "")[:400].strip()
                 if excerpt:
                     lines.append(f"    > {excerpt}")
@@ -793,6 +809,7 @@ async def run_compilation(
     external_data_filtered = {
         "rss_items": filtered_rss,
         "weather": [],
+        "topic_scores": topic_scores,
     }
 
     briefing_text = await _llm_synthesise(