Update src/api/agent.py

src/api/agent.py

@@ -1224,16 +1224,49 @@ def _contact_footer_for_query(question: str, context: Optional[SearchContextPayl
     return footer
 
 
+def _thought_block(intent: str, tool: str, question: str, extra: str = "") -> str:
+    """
+    Lightweight collapsible reasoning block shown at the top of every non-search response.
+    """
+    labels = {
+        "library_info":    ("📋", "Library information lookup", "I searched the KU Library knowledge base for relevant content."),
+        "general":         ("🌐", "General knowledge / web search", "I answered using a web search for current information."),
+        "general_recent":  ("🔴", "Live web search", "I ran a live web search for the most up-to-date information."),
+        "social_greeting": ("🐝", "Conversational response", "I responded directly as LibBee, the KU Library AI Assistant."),
+        "social":          ("🐝", "Conversational response", "I responded directly as LibBee, the KU Library AI Assistant."),
+        "staff_lookup":    ("👤", "Staff directory lookup", "I matched your request to the KU Library staff directory."),
+        "hours":           ("⏰", "Library hours lookup", "I retrieved the current library hours from the KU Library knowledge base."),
+        "campus":          ("📍", "Campus information lookup", "I retrieved KU Library campus information from the knowledge base."),
+        "database_rec":    ("🗄️", "Database recommendation", "I matched your subject to the KU Library e-resources database."),
+        "rag_search":      ("📖", "Knowledge base search", "I ran a hybrid semantic + keyword search across the KU Library knowledge base."),
+    }
+    icon, label, reasoning = labels.get(intent, ("🤖", "AI response", "I processed your question and generated a response."))
+    if extra:
+        reasoning += " " + extra
+
+    return (
+        f'<details style="margin:0 0 12px 0;border:1px solid #e5e7eb;border-radius:10px;'
+        f'background:#fafafa;overflow:hidden">'
+        f'<summary style="padding:8px 12px;cursor:pointer;font-size:.75rem;font-weight:700;'
+        f'color:#6b7280;list-style:none;display:flex;align-items:center;gap:5px">'
+        f'<span>{icon}</span> My approach — click to see</summary>'
+        f'<div style="padding:4px 12px 10px 12px;font-size:.79rem;line-height:1.7;border-top:1px solid #e5e7eb">'
+        f'<strong>Question:</strong> {_escape(question)}<br>'
+        f'<strong>Classified as:</strong> {label}<br>'
+        f'<strong>How I answered:</strong> {reasoning}'
+        f'</div></details>'
+    )
+
+
 def _search_trace_block(question: str, context: SearchContextPayload) -> str:
     """
-    'How I searched' reasoning panel — shown at the top of every search response.
-    Collapsible <details> so it doesn't dominate the answer.
+    Richer collapsible reasoning block for search responses.
+    Shows topic, boolean query, and reasoning. Also hints that AI tools are at the end.
     """
     topic = _escape(context.display_topic or context.topic)
     primo_boolean = _escape(context.primo_boolean_query or "")
     source_label = "PubMed" if context.source == "pubmed" else "KU PRIMO Library Discovery"
 
-    # Build reasoning sentence
     intent_label = {
         "search_academic": "academic literature search",
         "search_medical": "medical / clinical literature search",
@@ -1255,9 +1288,9 @@ def _search_trace_block(question: str, context: SearchContextPayload) -> str:
     filters_str = ", ".join(filters_parts)
 
     reasoning = (
-        f"I identified this as a <strong>{intent_label}</strong>. "
-        f"I extracted the topic <strong>{topic}</strong>, "
-        f"built an LLM-optimised boolean query with quoted phrases and synonyms, "
+        f"I classified this as a <strong>{intent_label}</strong>, "
+        f"extracted the topic <strong>{topic}</strong>, "
+        f"built an LLM-optimised boolean query with quoted phrases and OR synonyms, "
         f"then searched <strong>{source_label}</strong> for <strong>{filters_str}</strong>."
     )
     if context.intent == "search_medical" and context.source != "pubmed":
@@ -1278,7 +1311,8 @@ def _search_trace_block(question: str, context: SearchContextPayload) -> str:
             f'<code style="display:block;margin:4px 0;padding:6px 10px;background:#1a1a2e;color:#C8A951;'
             f'border-radius:7px;font-size:.78rem;word-break:break-all">{primo_boolean}</code>'
         )
-    block += f'<strong>Reasoning:</strong> {reasoning}'
+    block += f'<strong>Reasoning:</strong> {reasoning}<br>'
+    block += '<span style="color:#5b21b6;font-size:.78rem">✨ AI research tools for deeper exploration are listed at the end of this answer.</span>'
     block += '</div></details>'
     return block
 
@@ -2113,7 +2147,7 @@ async def agent_query(request: Request, req: AgentRequest):
         _safe_metrics_bucket("intents", "staff_lookup")
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=staff_name_answer(staff_match) + _source_badge("ku_library", "https://library.ku.ac.ae/"),
+            answer=_thought_block("staff_lookup", "staff_name_match", question) + staff_name_answer(staff_match) + _source_badge("ku_library", "https://library.ku.ac.ae/"),
             intent="library_info",
             tools_used=["staff_name_match"],
             model_used=model,
@@ -2127,7 +2161,7 @@ async def agent_query(request: Request, req: AgentRequest):
         _safe_metrics_bucket("intents", "staff_lookup")
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=staff_role_answer(staff_role_match, question) + _source_badge("ku_library", "https://library.ku.ac.ae/"),
+            answer=_thought_block("staff_lookup", "staff_role_match", question) + staff_role_answer(staff_role_match, question) + _source_badge("ku_library", "https://library.ku.ac.ae/"),
             intent="library_info",
             tools_used=["staff_role_match"],
             model_used=model,
@@ -2139,7 +2173,7 @@ async def agent_query(request: Request, req: AgentRequest):
     if _is_greeting_menu_followup(question, history):
         follow_up_question, follow_up_suggestions = _social_follow_up()
         return AgentResponse(
-            answer=_greeting_menu_clarify_answer() + _source_badge("libbee"),
+            answer=_thought_block("social_greeting", "greeting_menu", question) + _greeting_menu_clarify_answer() + _source_badge("libbee"),
             intent="social_greeting",
             is_follow_up=True,
             model_used=model,
@@ -2151,7 +2185,7 @@ async def agent_query(request: Request, req: AgentRequest):
     if _looks_library_hours_question(question):
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=_hours_answer() + _source_badge("ku_library", LIBRARY_HOURS_URL),
+            answer=_thought_block("hours", "hours_hardcoded", question) + _hours_answer() + _source_badge("ku_library", LIBRARY_HOURS_URL),
             intent="library_info",
             tools_used=["hours_hardcoded"],
             ask_librarian=False,
@@ -2164,7 +2198,7 @@ async def agent_query(request: Request, req: AgentRequest):
     if _looks_campus_question(question):
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=_campus_answer() + _source_badge("ku_library", "https://library.ku.ac.ae/"),
+            answer=_thought_block("campus", "campus_hardcoded", question) + _campus_answer() + _source_badge("ku_library", "https://library.ku.ac.ae/"),
             intent="library_info",
             tools_used=["campus_hardcoded"],
             ask_librarian=False,
@@ -2180,7 +2214,7 @@ async def agent_query(request: Request, req: AgentRequest):
         answer += _source_badge("ku_library", "https://library.ku.ac.ae/eresources")
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=answer,
+            answer=_thought_block("database_rec", "database_recommendation", question) + answer,
             intent="library_info",
             tools_used=["database_recommendation"],
             model_used=model,
@@ -2200,10 +2234,13 @@ async def agent_query(request: Request, req: AgentRequest):
             if follow_up_action == "summarize_topic":
                 answer, search_results, sources = await _research_snapshot(context, model)
             else:
+                # deep_research_tools / best_databases follow-up
                 answer = (
+                    f"{_search_trace_block(question, context)}"
                     f"<strong>Research guide for {_escape(context.display_topic or context.topic)}</strong><br><br>"
-                    f"{_database_guidance_for_topic(context)}"
+                    f"Here are the best AI tools and resources for a deep dive on this topic."
                     + _source_badge("libbee")
+                    + _ai_tools_footer(context)
                 )
                 search_results = []
                 sources = []
@@ -2269,7 +2306,7 @@ async def agent_query(request: Request, req: AgentRequest):
         )
         follow_up_question, follow_up_suggestions = _social_follow_up()
         return AgentResponse(
-            answer=answer + _source_badge("libbee"),
+            answer=_thought_block("social", "libbee_casual", question) + answer + _source_badge("libbee"),
             intent="social_greeting",
             tools_used=["libbee_casual"],
             model_used=model,
@@ -2279,7 +2316,6 @@ async def agent_query(request: Request, req: AgentRequest):
         )
 
     if intent == "library_info":
-        # FIX: use get_rag_service() accessor — safe against circular import failures
         rag_results = []
         try:
             from app import get_rag_service
@@ -2297,19 +2333,22 @@ async def agent_query(request: Request, req: AgentRequest):
             answer += _source_badge("ku_library", ASK_LIBRARIAN_URL)
             sources = []
             tools_used = ["rag_no_results"]
+            thought = _thought_block("library_info", "rag_no_results", question,
+                                     "No matching content was found — I am directing you to Ask a Librarian.")
         else:
             answer = await _rag_answer(question, rag_results, history, model)
             answer += _source_badge("rag", "https://library.ku.ac.ae/")
             sources = [
                 {"title": r.get("title", "Library KB"), "url": r.get("source", "")}
-                for r in rag_results
-                if r.get("source")
+                for r in rag_results if r.get("source")
             ]
             tools_used = ["rag_hybrid_search"]
+            thought = _thought_block("rag_search", "rag_hybrid_search", question,
+                                     f"Hybrid FAISS + BM25 search returned {len(rag_results)} relevant chunk(s) from the KB.")
 
         follow_up_question, follow_up_suggestions = _library_follow_up(question)
         return AgentResponse(
-            answer=answer,
+            answer=thought + answer,
             intent="library_info",
             tools_used=tools_used,
             sources=sources,
@@ -2346,8 +2385,10 @@ async def agent_query(request: Request, req: AgentRequest):
 
         if summary_mode or deep_research_mode:
             answer, search_results, sources = await _research_snapshot(context, model)
-            if deep_research_mode and "Research guide" not in answer:
-                answer += f"<br><br><strong>Research guide:</strong> {_database_guidance_for_topic(context)}"
+            # _research_snapshot already includes _search_trace_block and _ai_tools_footer
+            # Only append extra deep-research guidance if not already present
+            if deep_research_mode and "Research guide" not in answer and "AI tools" not in answer:
+                answer += f"<br><br>{_ai_tools_footer(context)}"
             follow_up_question, follow_up_suggestions = await _search_follow_up(context, model, summary_mode=True)
             _safe_metrics_increment("search_handoffs")
             return AgentResponse(
@@ -2398,7 +2439,7 @@ async def agent_query(request: Request, req: AgentRequest):
         answer = await _web_search_answer(question, history, model) + _source_badge("web_live")
         follow_up_question, follow_up_suggestions = _general_follow_up(question)
         return AgentResponse(
-            answer=answer,
+            answer=_thought_block("general_recent", "web_search", question) + answer,
             intent="general_recent",
             tools_used=["web_search"],
             model_used=model,
@@ -2407,10 +2448,11 @@ async def agent_query(request: Request, req: AgentRequest):
             follow_up_suggestions=follow_up_suggestions,
         )
 
+    # general fallback
     answer = await _web_search_answer(question, history, model) + _source_badge("web")
     follow_up_question, follow_up_suggestions = _general_follow_up(question)
     return AgentResponse(
-        answer=answer,
+        answer=_thought_block("general", "web_search", question) + answer,
         intent="general",
         tools_used=["web_search"],
         model_used=model,