fix: avoid over-rewriting query and wrong reference display

src/memos/memories/textual/tree_text_memory/retrieve/searcher.py

@@ -75,6 +75,18 @@ def search(
             context = [related_node["memory"] for related_node in related_nodes]
             context = list(set(context))
 
+            # add some knowledge retrieved from internet to the context to avoid misunderstanding while parsing the task goal.
+            if self.internet_retriever:
+                supplyment_memory_items = self.internet_retriever.retrieve_from_internet(
+                    query=query, top_k=3
+                )
+                context.extend(
+                    [
+                        each_supplyment_item.memory.partition("\nContent: ")[-1]
+                        for each_supplyment_item in supplyment_memory_items
+                    ]
+                )
+
         # Step 1a: Parse task structure into topic, concept, and fact levels
         parsed_goal = self.task_goal_parser.parse(
             task_description=query,

src/memos/memories/textual/tree_text_memory/retrieve/utils.py

@@ -5,7 +5,7 @@
 2. Tags: thematic tags to help categorize and retrieve related memories.
 3. Goal Type: retrieval | qa | generation
 4. Rephrased instruction: Give a rephrased task instruction based on the former conversation to make it less confusing to look alone. If you think the task instruction is easy enough to understand, or there is no former conversation, set "rephrased_instruction" to an empty string.
-5. Need for internet search: If you think you need to search the internet to finish the rephrased/original user task instruction, set "internet_search" to True. Otherwise, set it to False.
+5. Need for internet search: If the user's task instruction only involves objective facts or can be completed without introducing external knowledge, set "internet_search" to False. Otherwise, set it to True.
 6. Memories: Provide 2–5 short semantic expansions or rephrasings of the rephrased/original user task instruction. These are used for improved embedding search coverage. Each should be clear, concise, and meaningful for retrieval.
 
 Task description:

src/memos/templates/mos_prompts.py

@@ -92,7 +92,7 @@
   - `memoriesID` is the specific memory ID from the available memories list
 
 ### Reference Examples
-- Correct: `[1:abc123]`, `[2:def456]`, `[3:ghi789]`, `[4:jkl101]`, `[5:mno112]`
+- Correct: `[1:abc123]`, `[2:def456]`, `[3:ghi789]`, `[4:jkl101][5:mno112]` (concatenate reference annotation directly while citing multiple memories)
 - Incorrect: `[1:abc123,2:def456]` (do not use connected format)
 
 ## Response Guidelines