Wire chatbot search chunker and summy

api/chatbot.py

@@ -8,6 +8,7 @@ from .retrieval import retrieval_engine
 from memory import MemoryManager
 from utils import translate_query, process_medical_image
 from search import search_web
+from models import summarizer
 from models import process_search_query
 
 logger = logging.getLogger("medical-chatbot")
@@ -60,8 +61,19 @@ class RAGMedicalChatbot:
                 search_results = search_web(user_query, num_results=10)
                 if search_results:
                     logger.info(f"[SEARCH] Retrieved {len(search_results)} web resources")
-                    # Process with Llama
-                    search_context, url_mapping = process_search_query(user_query, search_results)
+                    # Compose a compact context strictly from query-relevant snippets
+                    # Also build URL mapping for citations
+                    url_mapping = {doc['id']: doc['url'] for doc in search_results if doc.get('id') and doc.get('url')}
+                    # Create per-doc short summaries (already relevant snippets), further compress
+                    summaries = []
+                    for doc in search_results:
+                        content = (doc.get('content') or '').strip()
+                        if not content:
+                            continue
+                        concise = summarizer.summarize_for_query(content, user_query, max_length=320)
+                        if concise:
+                            summaries.append(f"Document {doc['id']}: {concise}")
+                    search_context = "\n\n".join(summaries)
                     logger.info(f"[SEARCH] Processed with Llama, generated {len(url_mapping)} URL mappings")
                 else:
                     logger.warning("[SEARCH] No search results found")