diff --git a/surfsense_backend/app/agents/new_chat/tools/user_memory.py b/surfsense_backend/app/agents/new_chat/tools/user_memory.py
index 3cefa2b02..32230394e 100644
--- a/surfsense_backend/app/agents/new_chat/tools/user_memory.py
+++ b/surfsense_backend/app/agents/new_chat/tools/user_memory.py
@@ -158,46 +158,79 @@ def create_save_memory_tool(
Returns:
A dictionary with the save status and memory details
"""
- # Validate category
+ # Log at the very start
+ logger.info(f">>> SAVE_MEMORY TOOL CALLED: content='{content}', category='{category}'")
+ print(f">>> SAVE_MEMORY TOOL CALLED: content='{content}', category='{category}'")
+
+ # Normalize and validate category (LLMs may send uppercase)
+ category = category.lower() if category else "fact"
valid_categories = ["preference", "fact", "instruction", "context"]
if category not in valid_categories:
category = "fact"
try:
+ logger.info(f"save_memory called: user_id={user_id}, search_space_id={search_space_id}, content={content[:50]}...")
+
# Convert user_id to UUID
uuid_user_id = _to_uuid(user_id)
+ logger.info(f"UUID conversion successful: {uuid_user_id}")
# Check if we've hit the memory limit
memory_count = await get_user_memory_count(
db_session, user_id, search_space_id
)
+ logger.info(f"Current memory count: {memory_count}")
+
if memory_count >= MAX_MEMORIES_PER_USER:
# Delete oldest memory to make room
await delete_oldest_memory(db_session, user_id, search_space_id)
# Generate embedding for the memory
+ logger.info("Generating embedding...")
embedding = config.embedding_model_instance.embed(content)
+ logger.info(f"Embedding generated, type: {type(embedding)}, len: {len(embedding) if hasattr(embedding, '__len__') else 'N/A'}")
- # Map string category to enum
- category_enum = MemoryCategory(category)
+ # Convert numpy array to list of Python floats for PostgreSQL
+ import numpy as np
+ if isinstance(embedding, np.ndarray):
+ embedding_list = embedding.tolist()
+ else:
+ embedding_list = list(embedding)
- # Create new memory
- new_memory = UserMemory(
- user_id=uuid_user_id,
- search_space_id=search_space_id,
- memory_text=content,
- category=category_enum,
- embedding=embedding,
- updated_at=datetime.now(UTC),
+ # Create new memory using ORM with proper enum handling
+ # Use the enum's value attribute directly
+ from sqlalchemy import text as sql_text
+
+ now = datetime.now(UTC)
+
+ # Use raw SQL with proper parameter binding for asyncpg
+ insert_sql = sql_text("""
+ INSERT INTO user_memories (user_id, search_space_id, memory_text, category, embedding, updated_at, created_at)
+ VALUES (:user_id, :search_space_id, :memory_text, CAST(:category AS memorycategory), :embedding, :updated_at, :created_at)
+ RETURNING id
+ """)
+
+ result = await db_session.execute(
+ insert_sql,
+ {
+ "user_id": uuid_user_id,
+ "search_space_id": search_space_id,
+ "memory_text": content,
+ "category": category, # Already lowercase string
+ "embedding": str(embedding_list), # Convert to string format for pgvector
+ "updated_at": now,
+ "created_at": now,
+ }
)
-
- db_session.add(new_memory)
+ new_memory_id = result.scalar_one()
+
+ logger.info("Committing...")
await db_session.commit()
- await db_session.refresh(new_memory)
+ logger.info(f"Memory saved successfully with id: {new_memory_id}")
return {
"status": "saved",
- "memory_id": new_memory.id,
+ "memory_id": new_memory_id,
"memory_text": content,
"category": category,
"message": f"I'll remember: {content}",
@@ -205,6 +238,8 @@ def create_save_memory_tool(
except Exception as e:
logger.exception(f"Failed to save memory for user {user_id}: {e}")
+ # Rollback the session to clear any failed transaction state
+ await db_session.rollback()
return {
"status": "error",
"error": str(e),
@@ -260,10 +295,15 @@ def create_recall_memory_tool(
A dictionary containing relevant memories and formatted context
"""
top_k = min(max(top_k, 1), 20) # Clamp between 1 and 20
+
+ # Log at the very start
+ logger.info(f">>> RECALL_MEMORY TOOL CALLED: query='{query}', category='{category}', top_k={top_k}")
+ print(f">>> RECALL_MEMORY TOOL CALLED: query='{query}', category='{category}', top_k={top_k}")
try:
# Convert user_id to UUID
uuid_user_id = _to_uuid(user_id)
+ logger.info(f"Recall memory for user: {uuid_user_id}, search_space: {search_space_id}")
if query:
# Semantic search using embeddings
@@ -307,6 +347,8 @@ def create_recall_memory_tool(
result = await db_session.execute(stmt)
memories = result.scalars().all()
+
+ logger.info(f"Found {len(memories)} memories")
# Format memories for response
memory_list = [
@@ -318,8 +360,12 @@ def create_recall_memory_tool(
}
for m in memories
]
+
+ logger.info(f"Formatted memory list: {memory_list}")
formatted_context = format_memories_for_context(memory_list)
+
+ logger.info(f"Returning {len(memory_list)} memories")
return {
"status": "success",
@@ -329,6 +375,8 @@ def create_recall_memory_tool(
}
except Exception as e:
+ logger.exception(f"Failed to recall memories for user {user_id}: {e}")
+ await db_session.rollback()
return {
"status": "error",
"error": str(e),
diff --git a/surfsense_backend/app/db.py b/surfsense_backend/app/db.py
index c23b133e2..550f0a704 100644
--- a/surfsense_backend/app/db.py
+++ b/surfsense_backend/app/db.py
@@ -475,10 +475,11 @@ class ChatCommentMention(BaseModel, TimestampMixin):
class MemoryCategory(str, Enum):
"""Categories for user memories."""
- PREFERENCE = "preference" # User preferences (e.g., "prefers dark mode")
- FACT = "fact" # Facts about the user (e.g., "is a Python developer")
- INSTRUCTION = "instruction" # Standing instructions (e.g., "always respond in bullet points")
- CONTEXT = "context" # Contextual information (e.g., "working on project X")
+ # Using lowercase keys to match PostgreSQL enum values
+ preference = "preference" # User preferences (e.g., "prefers dark mode")
+ fact = "fact" # Facts about the user (e.g., "is a Python developer")
+ instruction = "instruction" # Standing instructions (e.g., "always respond in bullet points")
+ context = "context" # Contextual information (e.g., "working on project X")
class UserMemory(BaseModel, TimestampMixin):
@@ -510,7 +511,7 @@ class UserMemory(BaseModel, TimestampMixin):
category = Column(
SQLAlchemyEnum(MemoryCategory),
nullable=False,
- default=MemoryCategory.FACT,
+ default=MemoryCategory.fact,
)
# Vector embedding for semantic search
embedding = Column(Vector(config.embedding_model_instance.dimension))
diff --git a/surfsense_web/app/dashboard/[search_space_id]/new-chat/[[...chat_id]]/page.tsx b/surfsense_web/app/dashboard/[search_space_id]/new-chat/[[...chat_id]]/page.tsx
index 43c33ba5a..742ccc9b8 100644
--- a/surfsense_web/app/dashboard/[search_space_id]/new-chat/[[...chat_id]]/page.tsx
+++ b/surfsense_web/app/dashboard/[search_space_id]/new-chat/[[...chat_id]]/page.tsx
@@ -32,6 +32,7 @@ import { DisplayImageToolUI } from "@/components/tool-ui/display-image";
import { GeneratePodcastToolUI } from "@/components/tool-ui/generate-podcast";
import { LinkPreviewToolUI } from "@/components/tool-ui/link-preview";
import { ScrapeWebpageToolUI } from "@/components/tool-ui/scrape-webpage";
+import { SaveMemoryToolUI, RecallMemoryToolUI } from "@/components/tool-ui/user-memory";
// import { WriteTodosToolUI } from "@/components/tool-ui/write-todos";
import { getBearerToken } from "@/lib/auth-utils";
import { createAttachmentAdapter, extractAttachmentContent } from "@/lib/chat/attachment-adapter";
@@ -1056,6 +1057,8 @@ export default function NewChatPage() {
+
+
{/* Disabled for now */}