Tests pass

Author: abrookins

CLAUDE.md

@@ -65,7 +65,7 @@ Working Memory (Session-scoped)  →  Long-term Memory (Persistent)
 ```python
 # Correct - Use RedisVL queries
 from redisvl.query import VectorQuery, FilterQuery
-query = VectorQuery(vector=embedding, vector_field_name="embedding", return_fields=["text"])
+query = VectorQuery(vector=embedding, vector_field_name="vector", return_fields=["text"])
 
 # Avoid - Direct redis client searches
 # redis.ft().search(...)  # Don't do this

agent-memory-client/agent_memory_client/client.py

@@ -12,8 +12,8 @@
 from typing import Any, Literal
 
 import httpx
-import ulid
 from pydantic import BaseModel
+from ulid import ULID
 
 from .exceptions import MemoryClientError, MemoryServerError, MemoryValidationError
 from .filters import (
@@ -362,7 +362,7 @@ async def add_memories_to_working_memory(
         # Auto-generate IDs for memories that don't have them
         for memory in final_memories:
             if not memory.id:
-                memory.id = str(ulid.new())
+                memory.id = str(ULID())
 
         # Create new working memory with the memories
         working_memory = WorkingMemory(

agent-memory-client/agent_memory_client/models.py

@@ -9,8 +9,8 @@
 from enum import Enum
 from typing import Any, Literal
 
-import ulid
 from pydantic import BaseModel, Field
+from ulid import ULID
 
 # Model name literals for model-specific window sizes
 ModelNameLiteral = Literal[
@@ -122,7 +122,7 @@ class ClientMemoryRecord(MemoryRecord):
     """A memory record with a client-provided ID"""
 
     id: str = Field(
-        default_factory=lambda: str(ulid.new()),
+        default_factory=lambda: str(ULID()),
         description="Client-provided ID generated by the client (ULID)",
     )
 

agent-memory-client/pyproject.toml

@@ -28,7 +28,7 @@ classifiers = [
 dependencies = [
     "httpx>=0.25.0",
     "pydantic>=2.0.0",
-    "ulid-py>=1.1.0",
+    "python-ulid>=3.0.0",
 ]
 
 [project.optional-dependencies]

agent-memory-client/uv.lock

@@ -1,8 +1,8 @@
 import tiktoken
-import ulid
 from fastapi import APIRouter, Depends, HTTPException
 from mcp.server.fastmcp.prompts import base
 from mcp.types import TextContent
+from ulid import ULID
 
 from agent_memory_server import long_term_memory, working_memory
 from agent_memory_server.auth import UserInfo, get_current_user
@@ -279,7 +279,7 @@ async def put_session_memory(
 
             memories = [
                 MemoryRecord(
-                    id=str(ulid.new()),
+                    id=str(ULID()),
                     session_id=session_id,
                     text=f"{msg.role}: {msg.content}",
                     namespace=updated_memory.namespace,
@@ -539,6 +539,16 @@ async def memory_prompt(
                     ),
                 )
             )
+        else:
+            # Always include a system message about long-term memories, even if empty
+            _messages.append(
+                SystemMessage(
+                    content=TextContent(
+                        type="text",
+                        text="## Long term memories related to the user's query\n No relevant long-term memories found.",
+                    ),
+                )
+            )
 
     _messages.append(
         base.UserMessage(

agent_memory_server/api.py

@@ -98,8 +98,8 @@ class Settings(BaseSettings):
     # RedisVL Settings (kept for backwards compatibility)
     redisvl_distance_metric: str = "COSINE"
     redisvl_vector_dimensions: str = "1536"
-    redisvl_index_name: str = "memory"
-    redisvl_index_prefix: str = "memory"
+    redisvl_index_name: str = "memory_idx"
+    redisvl_index_prefix: str = "memory_idx"
 
     # Docket settings
     docket_name: str = "memory-server"

agent_memory_server/config.py

@@ -5,11 +5,11 @@
 from datetime import UTC, datetime
 from typing import Any
 
-import ulid
 from redis.asyncio import Redis
 from redis.commands.search.query import Query
 from redisvl.query import VectorRangeQuery
 from redisvl.utils.vectorize import OpenAITextVectorizer
+from ulid import ULID
 
 from agent_memory_server.config import settings
 from agent_memory_server.dependencies import get_background_tasks
@@ -244,7 +244,7 @@ async def merge_memories_with_llm(memories: list[dict], llm_client: Any = None)
     # Create the merged memory
     merged_memory = {
         "text": merged_text.strip(),
-        "id_": str(ulid.new()),
+        "id_": str(ULID()),
         "user_id": user_id,
         "session_id": session_id,
         "namespace": namespace,
@@ -666,7 +666,7 @@ async def index_long_term_memories(
 
     # Schedule background tasks for topic/entity extraction
     for memory in processed_memories:
-        memory_id = memory.id or str(ulid.new())
+        memory_id = memory.id or str(ULID())
         await background_tasks.add_task(
             extract_memory_structure, memory_id, memory.text, memory.namespace
         )
@@ -946,7 +946,7 @@ async def count_long_term_memories(
         namespace: Optional namespace filter
         user_id: Optional user ID filter
         session_id: Optional session ID filter
-        redis_client: Optional Redis client (kept for compatibility)
+        redis_client: Optional Redis client (for compatibility - not used by adapter)
 
     Returns:
         Total count of memories matching filters
@@ -1230,7 +1230,7 @@ async def deduplicate_by_semantic_search(
 
             # Convert back to LongTermMemory
             merged_memory_obj = MemoryRecord(
-                id=memory.id or str(ulid.new()),
+                id=memory.id or str(ULID()),
                 text=merged_memory["text"],
                 user_id=merged_memory["user_id"],
                 session_id=merged_memory["session_id"],
@@ -1450,7 +1450,7 @@ async def extract_memories_from_messages(
 
                     # Create a new memory record from the extraction
                     extracted_memory = MemoryRecord(
-                        id=str(ulid.new()),  # Server-generated ID
+                        id=str(ULID()),  # Server-generated ID
                         text=memory_data["text"],
                         memory_type=memory_data.get("type", "semantic"),
                         topics=memory_data.get("topics", []),

agent_memory_server/long_term_memory.py

@@ -3,9 +3,9 @@
 from enum import Enum
 from typing import Literal
 
-import ulid
 from mcp.server.fastmcp.prompts import base
 from pydantic import BaseModel, Field
+from ulid import ULID
 
 from agent_memory_server.config import settings
 from agent_memory_server.filters import (
@@ -143,7 +143,7 @@ class ClientMemoryRecord(MemoryRecord):
     """A memory record with a client-provided ID"""
 
     id: str = Field(
-        default_factory=lambda: str(ulid.new()),
+        default_factory=lambda: str(ULID()),
         description="Client-provided ID for deduplication and overwrites",
     )
 
@@ -383,4 +383,4 @@ class MemoryPromptResponse(BaseModel):
 class LenientMemoryRecord(MemoryRecord):
     """A memory record that can be created without an ID"""
 
-    id: str | None = Field(default_factory=lambda: str(ulid.new()))
+    id: str | None = Field(default_factory=lambda: str(ULID()))

agent_memory_server/models.py

@@ -41,7 +41,8 @@ def get_search_index(
     distance_metric: str = settings.redisvl_distance_metric,
 ) -> AsyncSearchIndex:
     global _index
-    if _index is None:
+    # Check if we need to create a new index (no cached index or different Redis client)
+    if _index is None or _index._redis_client != redis:
         schema = {
             "index": {
                 "name": index_name,