agno 0.1.2__py3-none-any.whl → 2.3.13__py3-none-any.whl

agno/db/mysql/mysql.py

@@ -0,0 +1,2923 @@
+import time
+from datetime import date, datetime, timedelta, timezone
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Sequence, Tuple, Union
+from uuid import uuid4
+
+if TYPE_CHECKING:
+    from agno.tracing.schemas import Span, Trace
+
+from agno.db.base import BaseDb, SessionType
+from agno.db.migrations.manager import MigrationManager
+from agno.db.mysql.schemas import get_table_schema_definition
+from agno.db.mysql.utils import (
+    apply_sorting,
+    bulk_upsert_metrics,
+    calculate_date_metrics,
+    create_schema,
+    deserialize_cultural_knowledge_from_db,
+    fetch_all_sessions_data,
+    get_dates_to_calculate_metrics_for,
+    is_table_available,
+    is_valid_table,
+    serialize_cultural_knowledge_for_db,
+)
+from agno.db.schemas.culture import CulturalKnowledge
+from agno.db.schemas.evals import EvalFilterType, EvalRunRecord, EvalType
+from agno.db.schemas.knowledge import KnowledgeRow
+from agno.db.schemas.memory import UserMemory
+from agno.session import AgentSession, Session, TeamSession, WorkflowSession
+from agno.utils.log import log_debug, log_error, log_info, log_warning
+from agno.utils.string import generate_id
+
+try:
+    from sqlalchemy import TEXT, ForeignKey, Index, UniqueConstraint, and_, cast, func, update
+    from sqlalchemy.dialects import mysql
+    from sqlalchemy.engine import Engine, create_engine
+    from sqlalchemy.orm import scoped_session, sessionmaker
+    from sqlalchemy.schema import Column, MetaData, Table
+    from sqlalchemy.sql.expression import select, text
+except ImportError:
+    raise ImportError("`sqlalchemy` not installed. Please install it using `pip install sqlalchemy`")
+
+
+class MySQLDb(BaseDb):
+    def __init__(
+        self,
+        id: Optional[str] = None,
+        db_engine: Optional[Engine] = None,
+        db_schema: Optional[str] = None,
+        db_url: Optional[str] = None,
+        session_table: Optional[str] = None,
+        culture_table: Optional[str] = None,
+        memory_table: Optional[str] = None,
+        metrics_table: Optional[str] = None,
+        eval_table: Optional[str] = None,
+        knowledge_table: Optional[str] = None,
+        traces_table: Optional[str] = None,
+        spans_table: Optional[str] = None,
+        versions_table: Optional[str] = None,
+        create_schema: bool = True,
+    ):
+        """
+        Interface for interacting with a MySQL database.
+
+        The following order is used to determine the database connection:
+            1. Use the db_engine if provided
+            2. Use the db_url
+            3. Raise an error if neither is provided
+
+        Args:
+            id (Optional[str]): ID of the database.
+            db_url (Optional[str]): The database URL to connect to.
+            db_engine (Optional[Engine]): The SQLAlchemy database engine to use.
+            db_schema (Optional[str]): The database schema to use.
+            session_table (Optional[str]): Name of the table to store Agent, Team and Workflow sessions.
+            culture_table (Optional[str]): Name of the table to store cultural knowledge.
+            memory_table (Optional[str]): Name of the table to store memories.
+            metrics_table (Optional[str]): Name of the table to store metrics.
+            eval_table (Optional[str]): Name of the table to store evaluation runs data.
+            knowledge_table (Optional[str]): Name of the table to store knowledge content.
+            traces_table (Optional[str]): Name of the table to store run traces.
+            spans_table (Optional[str]): Name of the table to store span events.
+            versions_table (Optional[str]): Name of the table to store schema versions.
+            create_schema (bool): Whether to automatically create the database schema if it doesn't exist.
+                Set to False if schema is managed externally (e.g., via migrations). Defaults to True.
+
+        Raises:
+            ValueError: If neither db_url nor db_engine is provided.
+            ValueError: If none of the tables are provided.
+        """
+        if id is None:
+            base_seed = db_url or str(db_engine.url)  # type: ignore
+            schema_suffix = db_schema if db_schema is not None else "ai"
+            seed = f"{base_seed}#{schema_suffix}"
+            id = generate_id(seed)
+
+        super().__init__(
+            id=id,
+            session_table=session_table,
+            culture_table=culture_table,
+            memory_table=memory_table,
+            metrics_table=metrics_table,
+            eval_table=eval_table,
+            knowledge_table=knowledge_table,
+            traces_table=traces_table,
+            spans_table=spans_table,
+            versions_table=versions_table,
+        )
+
+        _engine: Optional[Engine] = db_engine
+        if _engine is None and db_url is not None:
+            _engine = create_engine(db_url)
+        if _engine is None:
+            raise ValueError("One of db_url or db_engine must be provided")
+
+        self.db_url: Optional[str] = db_url
+        self.db_engine: Engine = _engine
+        self.db_schema: str = db_schema if db_schema is not None else "ai"
+        self.metadata: MetaData = MetaData(schema=self.db_schema)
+        self.create_schema: bool = create_schema
+
+        # Initialize database session
+        self.Session: scoped_session = scoped_session(sessionmaker(bind=self.db_engine))
+
+    # -- DB methods --
+    def table_exists(self, table_name: str) -> bool:
+        """Check if a table with the given name exists in the MySQL database.
+
+        Args:
+            table_name: Name of the table to check
+
+        Returns:
+            bool: True if the table exists in the database, False otherwise
+        """
+        with self.Session() as sess:
+            return is_table_available(session=sess, table_name=table_name, db_schema=self.db_schema)
+
+    def _create_table(self, table_name: str, table_type: str) -> Table:
+        """
+        Create a table with the appropriate schema based on the table type.
+
+        Args:
+            table_name (str): Name of the table to create
+            table_type (str): Type of table (used to get schema definition)
+
+        Returns:
+            Table: SQLAlchemy Table object
+        """
+        try:
+            table_schema = get_table_schema_definition(table_type).copy()
+
+            columns: List[Column] = []
+            indexes: List[str] = []
+            unique_constraints: List[str] = []
+            schema_unique_constraints = table_schema.pop("_unique_constraints", [])
+
+            # Get the columns, indexes, and unique constraints from the table schema
+            for col_name, col_config in table_schema.items():
+                column_args = [col_name, col_config["type"]()]
+                column_kwargs = {}
+                if col_config.get("primary_key", False):
+                    column_kwargs["primary_key"] = True
+                if "nullable" in col_config:
+                    column_kwargs["nullable"] = col_config["nullable"]
+                if col_config.get("index", False):
+                    indexes.append(col_name)
+                if col_config.get("unique", False):
+                    column_kwargs["unique"] = True
+                    unique_constraints.append(col_name)
+
+                # Handle foreign key constraint
+                if "foreign_key" in col_config:
+                    fk_ref = col_config["foreign_key"]
+                    # For spans table, dynamically replace the traces table reference
+                    # with the actual trace table name configured for this db instance
+                    if table_type == "spans" and "trace_id" in fk_ref:
+                        fk_ref = f"{self.db_schema}.{self.trace_table_name}.trace_id"
+                    column_args.append(ForeignKey(fk_ref))
+
+                columns.append(Column(*column_args, **column_kwargs))  # type: ignore
+
+            # Create the table object
+            table = Table(table_name, self.metadata, *columns, schema=self.db_schema)
+
+            # Add multi-column unique constraints with table-specific names
+            for constraint in schema_unique_constraints:
+                constraint_name = f"{table_name}_{constraint['name']}"
+                constraint_columns = constraint["columns"]
+                table.append_constraint(UniqueConstraint(*constraint_columns, name=constraint_name))
+
+            # Add indexes to the table definition
+            for idx_col in indexes:
+                idx_name = f"idx_{table_name}_{idx_col}"
+                table.append_constraint(Index(idx_name, idx_col))
+
+            if self.create_schema:
+                with self.Session() as sess, sess.begin():
+                    create_schema(session=sess, db_schema=self.db_schema)
+
+            # Create table
+            table_created = False
+            if not self.table_exists(table_name):
+                table.create(self.db_engine, checkfirst=True)
+                log_debug(f"Successfully created table '{table_name}'")
+                table_created = True
+            else:
+                log_debug(f"Table {self.db_schema}.{table_name} already exists, skipping creation")
+
+            # Create indexes
+            for idx in table.indexes:
+                try:
+                    # Check if index already exists
+                    with self.Session() as sess:
+                        exists_query = text(
+                            "SELECT 1 FROM information_schema.statistics WHERE table_schema = :schema "
+                            "AND table_name = :table_name AND index_name = :index_name"
+                        )
+                        exists = (
+                            sess.execute(
+                                exists_query,
+                                {"schema": self.db_schema, "table_name": table_name, "index_name": idx.name},
+                            ).scalar()
+                            is not None
+                        )
+                        if exists:
+                            log_debug(
+                                f"Index {idx.name} already exists in {self.db_schema}.{table_name}, skipping creation"
+                            )
+                            continue
+
+                    idx.create(self.db_engine)
+
+                    log_debug(f"Created index: {idx.name} for table {self.db_schema}.{table_name}")
+                except Exception as e:
+                    log_error(f"Error creating index {idx.name}: {e}")
+
+            # Store the schema version for the created table
+            if table_name != self.versions_table_name and table_created:
+                latest_schema_version = MigrationManager(self).latest_schema_version
+                self.upsert_schema_version(table_name=table_name, version=latest_schema_version.public)
+                log_info(
+                    f"Successfully stored version {latest_schema_version.public} in database for table {table_name}"
+                )
+
+            return table
+
+        except Exception as e:
+            log_error(f"Could not create table {self.db_schema}.{table_name}: {e}")
+            raise
+
+    def _create_all_tables(self):
+        """Create all tables for the database."""
+        tables_to_create = [
+            (self.session_table_name, "sessions"),
+            (self.memory_table_name, "memories"),
+            (self.metrics_table_name, "metrics"),
+            (self.eval_table_name, "evals"),
+            (self.knowledge_table_name, "knowledge"),
+            (self.culture_table_name, "culture"),
+            (self.trace_table_name, "traces"),
+            (self.span_table_name, "spans"),
+            (self.versions_table_name, "versions"),
+        ]
+
+        for table_name, table_type in tables_to_create:
+            self._get_or_create_table(table_name=table_name, table_type=table_type, create_table_if_not_found=True)
+
+    def _get_table(self, table_type: str, create_table_if_not_found: Optional[bool] = False) -> Optional[Table]:
+        if table_type == "sessions":
+            self.session_table = self._get_or_create_table(
+                table_name=self.session_table_name,
+                table_type="sessions",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.session_table
+
+        if table_type == "memories":
+            self.memory_table = self._get_or_create_table(
+                table_name=self.memory_table_name,
+                table_type="memories",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.memory_table
+
+        if table_type == "metrics":
+            self.metrics_table = self._get_or_create_table(
+                table_name=self.metrics_table_name,
+                table_type="metrics",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.metrics_table
+
+        if table_type == "evals":
+            self.eval_table = self._get_or_create_table(
+                table_name=self.eval_table_name,
+                table_type="evals",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.eval_table
+
+        if table_type == "knowledge":
+            self.knowledge_table = self._get_or_create_table(
+                table_name=self.knowledge_table_name,
+                table_type="knowledge",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.knowledge_table
+
+        if table_type == "culture":
+            self.culture_table = self._get_or_create_table(
+                table_name=self.culture_table_name,
+                table_type="culture",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.culture_table
+
+        if table_type == "versions":
+            self.versions_table = self._get_or_create_table(
+                table_name=self.versions_table_name,
+                table_type="versions",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.versions_table
+
+        if table_type == "traces":
+            self.traces_table = self._get_or_create_table(
+                table_name=self.trace_table_name,
+                table_type="traces",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.traces_table
+
+        if table_type == "spans":
+            # Ensure traces table exists first (spans has FK to traces)
+            if create_table_if_not_found:
+                self._get_table(table_type="traces", create_table_if_not_found=True)
+
+            self.spans_table = self._get_or_create_table(
+                table_name=self.span_table_name,
+                table_type="spans",
+                create_table_if_not_found=create_table_if_not_found,
+            )
+            return self.spans_table
+
+        raise ValueError(f"Unknown table type: {table_type}")
+
+    def _get_or_create_table(
+        self, table_name: str, table_type: str, create_table_if_not_found: Optional[bool] = False
+    ) -> Optional[Table]:
+        """
+        Check if the table exists and is valid, else create it.
+
+        Args:
+            table_name (str): Name of the table to get or create
+            table_type (str): Type of table (used to get schema definition)
+
+        Returns:
+            Table: SQLAlchemy Table object representing the schema.
+        """
+
+        with self.Session() as sess, sess.begin():
+            table_is_available = is_table_available(session=sess, table_name=table_name, db_schema=self.db_schema)
+
+        if not table_is_available:
+            if not create_table_if_not_found:
+                return None
+
+            created_table = self._create_table(table_name=table_name, table_type=table_type)
+
+            return created_table
+
+        if not is_valid_table(
+            db_engine=self.db_engine,
+            table_name=table_name,
+            table_type=table_type,
+            db_schema=self.db_schema,
+        ):
+            raise ValueError(f"Table {self.db_schema}.{table_name} has an invalid schema")
+
+        try:
+            table = Table(table_name, self.metadata, schema=self.db_schema, autoload_with=self.db_engine)
+            return table
+
+        except Exception as e:
+            log_error(f"Error loading existing table {self.db_schema}.{table_name}: {e}")
+            raise
+
+    def get_latest_schema_version(self, table_name: str) -> str:
+        """Get the latest version of the database schema."""
+        table = self._get_table(table_type="versions", create_table_if_not_found=True)
+        with self.Session() as sess:
+            # Latest version for the given table
+            stmt = select(table).where(table.c.table_name == table_name).order_by(table.c.version.desc()).limit(1)  # type: ignore
+            result = sess.execute(stmt).fetchone()
+            if result is None:
+                return "2.0.0"
+            version_dict = dict(result._mapping)
+            return version_dict.get("version") or "2.0.0"
+
+    def upsert_schema_version(self, table_name: str, version: str) -> None:
+        """Upsert the schema version into the database."""
+        table = self._get_table(table_type="versions", create_table_if_not_found=True)
+        if table is None:
+            return
+        current_datetime = datetime.now().isoformat()
+        with self.Session() as sess, sess.begin():
+            stmt = mysql.insert(table).values(  # type: ignore
+                table_name=table_name,
+                version=version,
+                created_at=current_datetime,  # Store as ISO format string
+                updated_at=current_datetime,
+            )
+            # Update version if table_name already exists
+            stmt = stmt.on_duplicate_key_update(
+                version=version,
+                created_at=current_datetime,
+                updated_at=current_datetime,
+            )
+            sess.execute(stmt)
+
+    # -- Session methods --
+    def delete_session(self, session_id: str) -> bool:
+        """
+        Delete a session from the database.
+
+        Args:
+            session_id (str): ID of the session to delete
+
+        Returns:
+            bool: True if the session was deleted, False otherwise.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return False
+
+            with self.Session() as sess, sess.begin():
+                delete_stmt = table.delete().where(table.c.session_id == session_id)
+                result = sess.execute(delete_stmt)
+                if result.rowcount == 0:
+                    log_debug(f"No session found to delete with session_id: {session_id} in table {table.name}")
+                    return False
+                else:
+                    log_debug(f"Successfully deleted session with session_id: {session_id} in table {table.name}")
+                    return True
+
+        except Exception as e:
+            log_error(f"Error deleting session: {e}")
+            return False
+
+    def delete_sessions(self, session_ids: List[str]) -> None:
+        """Delete all given sessions from the database.
+        Can handle multiple session types in the same run.
+
+        Args:
+            session_ids (List[str]): The IDs of the sessions to delete.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                delete_stmt = table.delete().where(table.c.session_id.in_(session_ids))
+                result = sess.execute(delete_stmt)
+
+            log_debug(f"Successfully deleted {result.rowcount} sessions")
+
+        except Exception as e:
+            log_error(f"Error deleting sessions: {e}")
+
+    def get_session(
+        self,
+        session_id: str,
+        session_type: SessionType,
+        user_id: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Optional[Union[Session, Dict[str, Any]]]:
+        """
+        Read a session from the database.
+
+        Args:
+            session_id (str): ID of the session to read.
+            session_type (SessionType): Type of session to get.
+            user_id (Optional[str]): User ID to filter by. Defaults to None.
+            deserialize (Optional[bool]): Whether to serialize the session. Defaults to True.
+
+        Returns:
+            Union[Session, Dict[str, Any], None]:
+                - When deserialize=True: Session object
+                - When deserialize=False: Session dictionary
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return None
+
+            with self.Session() as sess:
+                stmt = select(table).where(table.c.session_id == session_id)
+
+                if user_id is not None:
+                    stmt = stmt.where(table.c.user_id == user_id)
+                result = sess.execute(stmt).fetchone()
+                if result is None:
+                    return None
+
+                session = dict(result._mapping)
+
+            if not deserialize:
+                return session
+
+            if session_type == SessionType.AGENT:
+                return AgentSession.from_dict(session)
+            elif session_type == SessionType.TEAM:
+                return TeamSession.from_dict(session)
+            elif session_type == SessionType.WORKFLOW:
+                return WorkflowSession.from_dict(session)
+            else:
+                raise ValueError(f"Invalid session type: {session_type}")
+
+        except Exception as e:
+            log_error(f"Exception reading from session table: {e}")
+            return None
+
+    def get_sessions(
+        self,
+        session_type: Optional[SessionType] = None,
+        user_id: Optional[str] = None,
+        component_id: Optional[str] = None,
+        session_name: Optional[str] = None,
+        start_timestamp: Optional[int] = None,
+        end_timestamp: Optional[int] = None,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[Session], Tuple[List[Dict[str, Any]], int]]:
+        """
+        Get all sessions in the given table. Can filter by user_id and entity_id.
+
+        Args:
+            session_type (Optional[SessionType]): The type of sessions to get.
+            user_id (Optional[str]): The ID of the user to filter by.
+            component_id (Optional[str]): The ID of the agent / workflow to filter by.
+            start_timestamp (Optional[int]): The start timestamp to filter by.
+            end_timestamp (Optional[int]): The end timestamp to filter by.
+            session_name (Optional[str]): The name of the session to filter by.
+            limit (Optional[int]): The maximum number of sessions to return. Defaults to None.
+            page (Optional[int]): The page number to return. Defaults to None.
+            sort_by (Optional[str]): The field to sort by. Defaults to None.
+            sort_order (Optional[str]): The sort order. Defaults to None.
+            deserialize (Optional[bool]): Whether to serialize the sessions. Defaults to True.
+
+        Returns:
+            Union[List[Session], Tuple[List[Dict], int]]:
+                - When deserialize=True: List of Session objects
+                - When deserialize=False: Tuple of (session dictionaries, total count)
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return [] if deserialize else ([], 0)
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+
+                # Filtering
+                if user_id is not None:
+                    stmt = stmt.where(table.c.user_id == user_id)
+                if component_id is not None:
+                    if session_type == SessionType.AGENT:
+                        stmt = stmt.where(table.c.agent_id == component_id)
+                    elif session_type == SessionType.TEAM:
+                        stmt = stmt.where(table.c.team_id == component_id)
+                    elif session_type == SessionType.WORKFLOW:
+                        stmt = stmt.where(table.c.workflow_id == component_id)
+                if start_timestamp is not None:
+                    stmt = stmt.where(table.c.created_at >= start_timestamp)
+                if end_timestamp is not None:
+                    stmt = stmt.where(table.c.created_at <= end_timestamp)
+                if session_name is not None:
+                    # MySQL JSON extraction syntax
+                    stmt = stmt.where(
+                        func.coalesce(
+                            func.json_unquote(func.json_extract(table.c.session_data, "$.session_name")), ""
+                        ).ilike(f"%{session_name}%")
+                    )
+                if session_type is not None:
+                    session_type_value = session_type.value if isinstance(session_type, SessionType) else session_type
+                    stmt = stmt.where(table.c.session_type == session_type_value)
+
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Sorting
+                stmt = apply_sorting(stmt, table, sort_by, sort_order)
+
+                # Paginating
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [] if deserialize else ([], 0)
+
+                session_dicts = [dict(row._mapping) for row in result]
+                if not deserialize:
+                    return session_dicts, total_count
+
+                if session_type == SessionType.AGENT:
+                    return [AgentSession.from_dict(record) for record in session_dicts]  # type: ignore
+                elif session_type == SessionType.TEAM:
+                    return [TeamSession.from_dict(record) for record in session_dicts]  # type: ignore
+                elif session_type == SessionType.WORKFLOW:
+                    return [WorkflowSession.from_dict(record) for record in session_dicts]  # type: ignore
+                else:
+                    raise ValueError(f"Invalid session type: {session_type}")
+
+        except Exception as e:
+            log_error(f"Exception getting sessions: {e}")
+            raise e
+
+    def rename_session(
+        self, session_id: str, session_type: SessionType, session_name: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[Session, Dict[str, Any]]]:
+        """
+        Rename a session in the database.
+
+        Args:
+            session_id (str): The ID of the session to rename.
+            session_type (SessionType): The type of session to rename.
+            session_name (str): The new name for the session.
+            deserialize (Optional[bool]): Whether to serialize the session. Defaults to True.
+
+        Returns:
+            Optional[Union[Session, Dict[str, Any]]]:
+                - When deserialize=True: Session object
+                - When deserialize=False: Session dictionary
+
+        Raises:
+            Exception: If an error occurs during renaming.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                # MySQL JSON_SET syntax
+                stmt = (
+                    update(table)
+                    .where(table.c.session_id == session_id)
+                    .where(table.c.session_type == session_type.value)
+                    .values(session_data=func.json_set(table.c.session_data, "$.session_name", session_name))
+                )
+                sess.execute(stmt)
+
+                # Fetch the updated row
+                select_stmt = select(table).where(table.c.session_id == session_id)
+                result = sess.execute(select_stmt)
+                row = result.fetchone()
+                if not row:
+                    return None
+
+            session = dict(row._mapping)
+            if not deserialize:
+                return session
+
+            # Return the appropriate session type
+            if session_type == SessionType.AGENT:
+                return AgentSession.from_dict(session)
+            elif session_type == SessionType.TEAM:
+                return TeamSession.from_dict(session)
+            elif session_type == SessionType.WORKFLOW:
+                return WorkflowSession.from_dict(session)
+            else:
+                raise ValueError(f"Invalid session type: {session_type}")
+
+        except Exception as e:
+            log_error(f"Exception renaming session: {e}")
+            return None
+
+    def upsert_session(
+        self, session: Session, deserialize: Optional[bool] = True
+    ) -> Optional[Union[Session, Dict[str, Any]]]:
+        """
+        Insert or update a session in the database.
+
+        Args:
+            session (Session): The session data to upsert.
+            deserialize (Optional[bool]): Whether to deserialize the session. Defaults to True.
+
+        Returns:
+            Optional[Union[Session, Dict[str, Any]]]:
+                - When deserialize=True: Session object
+                - When deserialize=False: Session dictionary
+
+        Raises:
+            Exception: If an error occurs during upsert.
+        """
+        try:
+            table = self._get_table(table_type="sessions", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            session_dict = session.to_dict()
+
+            if isinstance(session, AgentSession):
+                with self.Session() as sess, sess.begin():
+                    stmt = mysql.insert(table).values(
+                        session_id=session_dict.get("session_id"),
+                        session_type=SessionType.AGENT.value,
+                        agent_id=session_dict.get("agent_id"),
+                        user_id=session_dict.get("user_id"),
+                        runs=session_dict.get("runs"),
+                        agent_data=session_dict.get("agent_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        created_at=session_dict.get("created_at"),
+                        updated_at=session_dict.get("created_at"),
+                    )
+                    stmt = stmt.on_duplicate_key_update(
+                        agent_id=session_dict.get("agent_id"),
+                        user_id=session_dict.get("user_id"),
+                        agent_data=session_dict.get("agent_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        runs=session_dict.get("runs"),
+                        updated_at=int(time.time()),
+                    )
+                    sess.execute(stmt)
+
+                    # Fetch the row
+                    select_stmt = select(table).where(table.c.session_id == session_dict.get("session_id"))
+                    result = sess.execute(select_stmt)
+                    row = result.fetchone()
+                    if not row:
+                        return None
+                    session_dict = dict(row._mapping)
+                    if session_dict is None or not deserialize:
+                        return session_dict
+                    return AgentSession.from_dict(session_dict)
+
+            elif isinstance(session, TeamSession):
+                with self.Session() as sess, sess.begin():
+                    stmt = mysql.insert(table).values(
+                        session_id=session_dict.get("session_id"),
+                        session_type=SessionType.TEAM.value,
+                        team_id=session_dict.get("team_id"),
+                        user_id=session_dict.get("user_id"),
+                        runs=session_dict.get("runs"),
+                        team_data=session_dict.get("team_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        created_at=session_dict.get("created_at"),
+                        updated_at=session_dict.get("created_at"),
+                    )
+                    stmt = stmt.on_duplicate_key_update(
+                        team_id=session_dict.get("team_id"),
+                        user_id=session_dict.get("user_id"),
+                        team_data=session_dict.get("team_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        runs=session_dict.get("runs"),
+                        updated_at=int(time.time()),
+                    )
+                    sess.execute(stmt)
+
+                    # Fetch the row
+                    select_stmt = select(table).where(table.c.session_id == session_dict.get("session_id"))
+                    result = sess.execute(select_stmt)
+                    row = result.fetchone()
+                    if not row:
+                        return None
+                    session_dict = dict(row._mapping)
+                    if session_dict is None or not deserialize:
+                        return session_dict
+                    return TeamSession.from_dict(session_dict)
+
+            else:
+                with self.Session() as sess, sess.begin():
+                    stmt = mysql.insert(table).values(
+                        session_id=session_dict.get("session_id"),
+                        session_type=SessionType.WORKFLOW.value,
+                        workflow_id=session_dict.get("workflow_id"),
+                        user_id=session_dict.get("user_id"),
+                        runs=session_dict.get("runs"),
+                        workflow_data=session_dict.get("workflow_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        created_at=session_dict.get("created_at"),
+                        updated_at=session_dict.get("created_at"),
+                    )
+                    stmt = stmt.on_duplicate_key_update(
+                        workflow_id=session_dict.get("workflow_id"),
+                        user_id=session_dict.get("user_id"),
+                        workflow_data=session_dict.get("workflow_data"),
+                        session_data=session_dict.get("session_data"),
+                        summary=session_dict.get("summary"),
+                        metadata=session_dict.get("metadata"),
+                        runs=session_dict.get("runs"),
+                        updated_at=int(time.time()),
+                    )
+                    sess.execute(stmt)
+
+                    # Fetch the row
+                    select_stmt = select(table).where(table.c.session_id == session_dict.get("session_id"))
+                    result = sess.execute(select_stmt)
+                    row = result.fetchone()
+                    if not row:
+                        return None
+                    session_dict = dict(row._mapping)
+                    if session_dict is None or not deserialize:
+                        return session_dict
+                    return WorkflowSession.from_dict(session_dict)
+
+        except Exception as e:
+            log_error(f"Exception upserting into sessions table: {e}")
+            return None
+
+    def upsert_sessions(
+        self, sessions: List[Session], deserialize: Optional[bool] = True, preserve_updated_at: bool = False
+    ) -> List[Union[Session, Dict[str, Any]]]:
+        """
+        Bulk upsert multiple sessions for improved performance on large datasets.
+
+        Args:
+            sessions (List[Session]): List of sessions to upsert.
+            deserialize (Optional[bool]): Whether to deserialize the sessions. Defaults to True.
+            preserve_updated_at (bool): If True, preserve the updated_at from the session object.
+
+        Returns:
+            List[Union[Session, Dict[str, Any]]]: List of upserted sessions.
+
+        Raises:
+            Exception: If an error occurs during bulk upsert.
+        """
+        if not sessions:
+            return []
+
+        try:
+            table = self._get_table(table_type="sessions", create_table_if_not_found=True)
+            if table is None:
+                log_info("Sessions table not available, falling back to individual upserts")
+                return [
+                    result
+                    for session in sessions
+                    if session is not None
+                    for result in [self.upsert_session(session, deserialize=deserialize)]
+                    if result is not None
+                ]
+
+            # Group sessions by type for batch processing
+            agent_sessions = []
+            team_sessions = []
+            workflow_sessions = []
+
+            for session in sessions:
+                if isinstance(session, AgentSession):
+                    agent_sessions.append(session)
+                elif isinstance(session, TeamSession):
+                    team_sessions.append(session)
+                elif isinstance(session, WorkflowSession):
+                    workflow_sessions.append(session)
+
+            results: List[Union[Session, Dict[str, Any]]] = []
+
+            # Process each session type in bulk
+            with self.Session() as sess, sess.begin():
+                # Bulk upsert agent sessions
+                if agent_sessions:
+                    agent_data = []
+                    for session in agent_sessions:
+                        session_dict = session.to_dict()
+                        # Use preserved updated_at if flag is set and value exists, otherwise use current time
+                        updated_at = session_dict.get("updated_at") if preserve_updated_at else int(time.time())
+                        agent_data.append(
+                            {
+                                "session_id": session_dict.get("session_id"),
+                                "session_type": SessionType.AGENT.value,
+                                "agent_id": session_dict.get("agent_id"),
+                                "user_id": session_dict.get("user_id"),
+                                "runs": session_dict.get("runs"),
+                                "agent_data": session_dict.get("agent_data"),
+                                "session_data": session_dict.get("session_data"),
+                                "summary": session_dict.get("summary"),
+                                "metadata": session_dict.get("metadata"),
+                                "created_at": session_dict.get("created_at"),
+                                "updated_at": updated_at,
+                            }
+                        )
+
+                    if agent_data:
+                        stmt = mysql.insert(table)
+                        stmt = stmt.on_duplicate_key_update(
+                            agent_id=stmt.inserted.agent_id,
+                            user_id=stmt.inserted.user_id,
+                            agent_data=stmt.inserted.agent_data,
+                            session_data=stmt.inserted.session_data,
+                            summary=stmt.inserted.summary,
+                            metadata=stmt.inserted.metadata,
+                            runs=stmt.inserted.runs,
+                            updated_at=stmt.inserted.updated_at,
+                        )
+                        sess.execute(stmt, agent_data)
+
+                        # Fetch the results for agent sessions
+                        agent_ids = [session.session_id for session in agent_sessions]
+                        select_stmt = select(table).where(table.c.session_id.in_(agent_ids))
+                        result = sess.execute(select_stmt).fetchall()
+
+                        for row in result:
+                            session_dict = dict(row._mapping)
+                            if deserialize:
+                                deserialized_agent_session = AgentSession.from_dict(session_dict)
+                                if deserialized_agent_session is None:
+                                    continue
+                                results.append(deserialized_agent_session)
+                            else:
+                                results.append(session_dict)
+
+                # Bulk upsert team sessions
+                if team_sessions:
+                    team_data = []
+                    for session in team_sessions:
+                        session_dict = session.to_dict()
+                        # Use preserved updated_at if flag is set and value exists, otherwise use current time
+                        updated_at = session_dict.get("updated_at") if preserve_updated_at else int(time.time())
+                        team_data.append(
+                            {
+                                "session_id": session_dict.get("session_id"),
+                                "session_type": SessionType.TEAM.value,
+                                "team_id": session_dict.get("team_id"),
+                                "user_id": session_dict.get("user_id"),
+                                "runs": session_dict.get("runs"),
+                                "team_data": session_dict.get("team_data"),
+                                "session_data": session_dict.get("session_data"),
+                                "summary": session_dict.get("summary"),
+                                "metadata": session_dict.get("metadata"),
+                                "created_at": session_dict.get("created_at"),
+                                "updated_at": updated_at,
+                            }
+                        )
+
+                    if team_data:
+                        stmt = mysql.insert(table)
+                        stmt = stmt.on_duplicate_key_update(
+                            team_id=stmt.inserted.team_id,
+                            user_id=stmt.inserted.user_id,
+                            team_data=stmt.inserted.team_data,
+                            session_data=stmt.inserted.session_data,
+                            summary=stmt.inserted.summary,
+                            metadata=stmt.inserted.metadata,
+                            runs=stmt.inserted.runs,
+                            updated_at=stmt.inserted.updated_at,
+                        )
+                        sess.execute(stmt, team_data)
+
+                        # Fetch the results for team sessions
+                        team_ids = [session.session_id for session in team_sessions]
+                        select_stmt = select(table).where(table.c.session_id.in_(team_ids))
+                        result = sess.execute(select_stmt).fetchall()
+
+                        for row in result:
+                            session_dict = dict(row._mapping)
+                            if deserialize:
+                                deserialized_team_session = TeamSession.from_dict(session_dict)
+                                if deserialized_team_session is None:
+                                    continue
+                                results.append(deserialized_team_session)
+                            else:
+                                results.append(session_dict)
+
+                # Bulk upsert workflow sessions
+                if workflow_sessions:
+                    workflow_data = []
+                    for session in workflow_sessions:
+                        session_dict = session.to_dict()
+                        # Use preserved updated_at if flag is set and value exists, otherwise use current time
+                        updated_at = session_dict.get("updated_at") if preserve_updated_at else int(time.time())
+                        workflow_data.append(
+                            {
+                                "session_id": session_dict.get("session_id"),
+                                "session_type": SessionType.WORKFLOW.value,
+                                "workflow_id": session_dict.get("workflow_id"),
+                                "user_id": session_dict.get("user_id"),
+                                "runs": session_dict.get("runs"),
+                                "workflow_data": session_dict.get("workflow_data"),
+                                "session_data": session_dict.get("session_data"),
+                                "summary": session_dict.get("summary"),
+                                "metadata": session_dict.get("metadata"),
+                                "created_at": session_dict.get("created_at"),
+                                "updated_at": updated_at,
+                            }
+                        )
+
+                    if workflow_data:
+                        stmt = mysql.insert(table)
+                        stmt = stmt.on_duplicate_key_update(
+                            workflow_id=stmt.inserted.workflow_id,
+                            user_id=stmt.inserted.user_id,
+                            workflow_data=stmt.inserted.workflow_data,
+                            session_data=stmt.inserted.session_data,
+                            summary=stmt.inserted.summary,
+                            metadata=stmt.inserted.metadata,
+                            runs=stmt.inserted.runs,
+                            updated_at=stmt.inserted.updated_at,
+                        )
+                        sess.execute(stmt, workflow_data)
+
+                        # Fetch the results for workflow sessions
+                        workflow_ids = [session.session_id for session in workflow_sessions]
+                        select_stmt = select(table).where(table.c.session_id.in_(workflow_ids))
+                        result = sess.execute(select_stmt).fetchall()
+
+                        for row in result:
+                            session_dict = dict(row._mapping)
+                            if deserialize:
+                                deserialized_workflow_session = WorkflowSession.from_dict(session_dict)
+                                if deserialized_workflow_session is None:
+                                    continue
+                                results.append(deserialized_workflow_session)
+                            else:
+                                results.append(session_dict)
+
+            return results
+
+        except Exception as e:
+            log_error(f"Exception during bulk session upsert, falling back to individual upserts: {e}")
+            # Fallback to individual upserts
+            return [
+                result
+                for session in sessions
+                if session is not None
+                for result in [self.upsert_session(session, deserialize=deserialize)]
+                if result is not None
+            ]
+
+    # -- Memory methods --
+    def delete_user_memory(self, memory_id: str, user_id: Optional[str] = None):
+        """Delete a user memory from the database.
+
+        Args:
+            memory_id (str): The ID of the memory to delete.
+            user_id (Optional[str]): The user ID to filter by. Defaults to None.
+
+        Returns:
+            bool: True if deletion was successful, False otherwise.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                delete_stmt = table.delete().where(table.c.memory_id == memory_id)
+                if user_id is not None:
+                    delete_stmt = delete_stmt.where(table.c.user_id == user_id)
+                result = sess.execute(delete_stmt)
+
+                success = result.rowcount > 0
+                if success:
+                    log_debug(f"Successfully deleted user memory id: {memory_id}")
+                else:
+                    log_debug(f"No user memory found with id: {memory_id}")
+
+        except Exception as e:
+            log_error(f"Error deleting user memory: {e}")
+
+    def delete_user_memories(self, memory_ids: List[str], user_id: Optional[str] = None) -> None:
+        """Delete user memories from the database.
+
+        Args:
+            memory_ids (List[str]): The IDs of the memories to delete.
+            user_id (Optional[str]): The user ID to filter by. Defaults to None.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                delete_stmt = table.delete().where(table.c.memory_id.in_(memory_ids))
+                if user_id is not None:
+                    delete_stmt = delete_stmt.where(table.c.user_id == user_id)
+                result = sess.execute(delete_stmt)
+                if result.rowcount == 0:
+                    log_debug(f"No user memories found with ids: {memory_ids}")
+
+        except Exception as e:
+            log_error(f"Error deleting user memories: {e}")
+
+    def get_all_memory_topics(self, user_id: Optional[str] = None) -> List[str]:
+        """Get all memory topics from the database.
+
+        Args:
+            user_id (Optional[str]): Optional user ID to filter topics.
+
+        Returns:
+            List[str]: List of memory topics.
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return []
+
+            with self.Session() as sess, sess.begin():
+                # MySQL approach: extract JSON array elements differently
+                stmt = select(table.c.topics)
+                result = sess.execute(stmt).fetchall()
+
+                topics_set = set()
+                for row in result:
+                    if row[0]:
+                        # Parse JSON array and add topics to set
+                        import json
+
+                        try:
+                            topics = json.loads(row[0]) if isinstance(row[0], str) else row[0]
+                            if isinstance(topics, list):
+                                topics_set.update(topics)
+                        except Exception:
+                            pass
+
+                return list(topics_set)
+
+        except Exception as e:
+            log_error(f"Exception reading from memory table: {e}")
+            raise e
+
+    def get_user_memory(
+        self, memory_id: str, deserialize: Optional[bool] = True, user_id: Optional[str] = None
+    ) -> Optional[UserMemory]:
+        """Get a memory from the database.
+
+        Args:
+            memory_id (str): The ID of the memory to get.
+            deserialize (Optional[bool]): Whether to serialize the memory. Defaults to True.
+            user_id (Optional[str]): The user ID to filter by. Defaults to None.
+
+        Returns:
+            Union[UserMemory, Dict[str, Any], None]:
+                - When deserialize=True: UserMemory object
+                - When deserialize=False: UserMemory dictionary
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table).where(table.c.memory_id == memory_id)
+                if user_id is not None:
+                    stmt = stmt.where(table.c.user_id == user_id)
+
+                result = sess.execute(stmt).fetchone()
+                if not result:
+                    return None
+
+                memory_raw = result._mapping
+                if not deserialize:
+                    return memory_raw
+            return UserMemory.from_dict(memory_raw)
+
+        except Exception as e:
+            log_error(f"Exception reading from memory table: {e}")
+            return None
+
+    def get_user_memories(
+        self,
+        user_id: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        topics: Optional[List[str]] = None,
+        search_content: Optional[str] = None,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[UserMemory], Tuple[List[Dict[str, Any]], int]]:
+        """Get all memories from the database as MemoryRow objects.
+
+        Args:
+            user_id (Optional[str]): The ID of the user to filter by.
+            agent_id (Optional[str]): The ID of the agent to filter by.
+            team_id (Optional[str]): The ID of the team to filter by.
+            topics (Optional[List[str]]): The topics to filter by.
+            search_content (Optional[str]): The content to search for.
+            limit (Optional[int]): The maximum number of memories to return.
+            page (Optional[int]): The page number.
+            sort_by (Optional[str]): The column to sort by.
+            sort_order (Optional[str]): The order to sort by.
+            deserialize (Optional[bool]): Whether to serialize the memories. Defaults to True.
+
+
+        Returns:
+            Union[List[UserMemory], Tuple[List[Dict[str, Any]], int]]:
+                - When deserialize=True: List of UserMemory objects
+                - When deserialize=False: Tuple of (memory dictionaries, total count)
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return [] if deserialize else ([], 0)
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+                # Filtering
+                if user_id is not None:
+                    stmt = stmt.where(table.c.user_id == user_id)
+                if agent_id is not None:
+                    stmt = stmt.where(table.c.agent_id == agent_id)
+                if team_id is not None:
+                    stmt = stmt.where(table.c.team_id == team_id)
+                if topics is not None:
+                    # MySQL JSON contains syntax
+                    topic_conditions = []
+                    for topic in topics:
+                        topic_conditions.append(func.json_contains(table.c.topics, f'"{topic}"'))
+                    stmt = stmt.where(and_(*topic_conditions))
+                if search_content is not None:
+                    stmt = stmt.where(cast(table.c.memory, TEXT).ilike(f"%{search_content}%"))
+
+                # Get total count after applying filtering
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Sorting
+                stmt = apply_sorting(stmt, table, sort_by, sort_order)
+
+                # Paginating
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [] if deserialize else ([], 0)
+
+                memories_raw = [record._mapping for record in result]
+                if not deserialize:
+                    return memories_raw, total_count
+
+            return [UserMemory.from_dict(record) for record in memories_raw]
+
+        except Exception as e:
+            log_error(f"Exception reading from memory table: {e}")
+            raise e
+
+    def clear_memories(self) -> None:
+        """Clear all user memories from the database."""
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                sess.execute(table.delete())
+        except Exception as e:
+            log_error(f"Exception clearing user memories: {e}")
+
+    def get_user_memory_stats(
+        self, limit: Optional[int] = None, page: Optional[int] = None, user_id: Optional[str] = None
+    ) -> Tuple[List[Dict[str, Any]], int]:
+        """Get user memories stats.
+
+        Args:
+            limit (Optional[int]): The maximum number of user stats to return.
+            page (Optional[int]): The page number.
+
+        Returns:
+            Tuple[List[Dict[str, Any]], int]: A list of dictionaries containing user stats and total count.
+
+        Example:
+        (
+            [
+                {
+                    "user_id": "123",
+                    "total_memories": 10,
+                    "last_memory_updated_at": 1714560000,
+                },
+            ],
+            total_count: 1,
+        )
+        """
+        try:
+            table = self._get_table(table_type="memories")
+            if table is None:
+                return [], 0
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(
+                    table.c.user_id,
+                    func.count(table.c.memory_id).label("total_memories"),
+                    func.max(table.c.updated_at).label("last_memory_updated_at"),
+                )
+
+                if user_id is not None:
+                    stmt = stmt.where(table.c.user_id == user_id)
+                else:
+                    stmt = stmt.where(table.c.user_id.is_not(None))
+
+                stmt = stmt.group_by(table.c.user_id)
+                stmt = stmt.order_by(func.max(table.c.updated_at).desc())
+
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Pagination
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [], 0
+
+                return [
+                    {
+                        "user_id": record.user_id,  # type: ignore
+                        "total_memories": record.total_memories,
+                        "last_memory_updated_at": record.last_memory_updated_at,
+                    }
+                    for record in result
+                ], total_count
+
+        except Exception as e:
+            log_error(f"Exception getting user memory stats: {e}")
+            return [], 0
+
+    def upsert_user_memory(
+        self, memory: UserMemory, deserialize: Optional[bool] = True
+    ) -> Optional[Union[UserMemory, Dict[str, Any]]]:
+        """Upsert a user memory in the database.
+
+        Args:
+            memory (UserMemory): The user memory to upsert.
+            deserialize (Optional[bool]): Whether to serialize the memory. Defaults to True.
+
+        Returns:
+            Optional[Union[UserMemory, Dict[str, Any]]]:
+                - When deserialize=True: UserMemory object
+                - When deserialize=False: UserMemory dictionary
+
+        Raises:
+            Exception: If an error occurs during upsert.
+        """
+        try:
+            table = self._get_table(table_type="memories", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                if memory.memory_id is None:
+                    memory.memory_id = str(uuid4())
+
+                current_time = int(time.time())
+
+                stmt = mysql.insert(table).values(
+                    memory_id=memory.memory_id,
+                    memory=memory.memory,
+                    input=memory.input,
+                    user_id=memory.user_id,
+                    agent_id=memory.agent_id,
+                    team_id=memory.team_id,
+                    topics=memory.topics,
+                    feedback=memory.feedback,
+                    created_at=memory.created_at,
+                    updated_at=memory.created_at,
+                )
+                stmt = stmt.on_duplicate_key_update(
+                    memory=memory.memory,
+                    topics=memory.topics,
+                    input=memory.input,
+                    agent_id=memory.agent_id,
+                    team_id=memory.team_id,
+                    feedback=memory.feedback,
+                    updated_at=current_time,
+                    # Preserve created_at on update - don't overwrite existing value
+                    created_at=table.c.created_at,
+                )
+                sess.execute(stmt)
+
+                # Fetch the row
+                select_stmt = select(table).where(table.c.memory_id == memory.memory_id)
+                result = sess.execute(select_stmt)
+                row = result.fetchone()
+                if not row:
+                    return None
+
+            memory_raw = row._mapping
+            if not memory_raw or not deserialize:
+                return memory_raw
+
+            return UserMemory.from_dict(memory_raw)
+
+        except Exception as e:
+            log_error(f"Exception upserting user memory: {e}")
+            return None
+
+    def upsert_memories(
+        self, memories: List[UserMemory], deserialize: Optional[bool] = True, preserve_updated_at: bool = False
+    ) -> List[Union[UserMemory, Dict[str, Any]]]:
+        """
+        Bulk upsert multiple user memories for improved performance on large datasets.
+
+        Args:
+            memories (List[UserMemory]): List of memories to upsert.
+            deserialize (Optional[bool]): Whether to deserialize the memories. Defaults to True.
+
+        Returns:
+            List[Union[UserMemory, Dict[str, Any]]]: List of upserted memories.
+
+        Raises:
+            Exception: If an error occurs during bulk upsert.
+        """
+        if not memories:
+            return []
+
+        try:
+            table = self._get_table(table_type="memories", create_table_if_not_found=True)
+            if table is None:
+                log_info("Memories table not available, falling back to individual upserts")
+                return [
+                    result
+                    for memory in memories
+                    if memory is not None
+                    for result in [self.upsert_user_memory(memory, deserialize=deserialize)]
+                    if result is not None
+                ]
+
+            # Prepare bulk data
+            bulk_data = []
+            current_time = int(time.time())
+
+            for memory in memories:
+                if memory.memory_id is None:
+                    memory.memory_id = str(uuid4())
+
+                # Use preserved updated_at if flag is set and value exists, otherwise use current time
+                updated_at = memory.updated_at if preserve_updated_at else current_time
+
+                bulk_data.append(
+                    {
+                        "memory_id": memory.memory_id,
+                        "memory": memory.memory,
+                        "input": memory.input,
+                        "user_id": memory.user_id,
+                        "agent_id": memory.agent_id,
+                        "team_id": memory.team_id,
+                        "topics": memory.topics,
+                        "feedback": memory.feedback,
+                        "created_at": memory.created_at,
+                        "updated_at": updated_at,
+                    }
+                )
+
+            results: List[Union[UserMemory, Dict[str, Any]]] = []
+
+            with self.Session() as sess, sess.begin():
+                # Bulk upsert memories using MySQL ON DUPLICATE KEY UPDATE
+                stmt = mysql.insert(table)
+                stmt = stmt.on_duplicate_key_update(
+                    memory=stmt.inserted.memory,
+                    topics=stmt.inserted.topics,
+                    input=stmt.inserted.input,
+                    agent_id=stmt.inserted.agent_id,
+                    team_id=stmt.inserted.team_id,
+                    feedback=stmt.inserted.feedback,
+                    updated_at=stmt.inserted.updated_at,
+                    # Preserve created_at on update
+                    created_at=table.c.created_at,
+                )
+                sess.execute(stmt, bulk_data)
+
+                # Fetch results
+                memory_ids = [memory.memory_id for memory in memories if memory.memory_id]
+                select_stmt = select(table).where(table.c.memory_id.in_(memory_ids))
+                result = sess.execute(select_stmt).fetchall()
+
+                for row in result:
+                    memory_dict = dict(row._mapping)
+                    if deserialize:
+                        results.append(UserMemory.from_dict(memory_dict))
+                    else:
+                        results.append(memory_dict)
+
+            return results
+
+        except Exception as e:
+            log_error(f"Exception during bulk memory upsert, falling back to individual upserts: {e}")
+            # Fallback to individual upserts
+            return [
+                result
+                for memory in memories
+                if memory is not None
+                for result in [self.upsert_user_memory(memory, deserialize=deserialize)]
+                if result is not None
+            ]
+
+    # -- Metrics methods --
+    def _get_all_sessions_for_metrics_calculation(
+        self, start_timestamp: Optional[int] = None, end_timestamp: Optional[int] = None
+    ) -> List[Dict[str, Any]]:
+        """
+        Get all sessions of all types (agent, team, workflow) as raw dictionaries.
+
+         Args:
+            start_timestamp (Optional[int]): The start timestamp to filter by. Defaults to None.
+            end_timestamp (Optional[int]): The end timestamp to filter by. Defaults to None.
+
+        Returns:
+            List[Dict[str, Any]]: List of session dictionaries with session_type field.
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="sessions")
+            if table is None:
+                return []
+
+            stmt = select(
+                table.c.user_id,
+                table.c.session_data,
+                table.c.runs,
+                table.c.created_at,
+                table.c.session_type,
+            )
+
+            if start_timestamp is not None:
+                stmt = stmt.where(table.c.created_at >= start_timestamp)
+            if end_timestamp is not None:
+                stmt = stmt.where(table.c.created_at <= end_timestamp)
+
+            with self.Session() as sess:
+                result = sess.execute(stmt).fetchall()
+                return [record._mapping for record in result]
+
+        except Exception as e:
+            log_error(f"Exception reading from sessions table: {e}")
+            raise e
+
+    def _get_metrics_calculation_starting_date(self, table: Table) -> Optional[date]:
+        """Get the first date for which metrics calculation is needed:
+
+        1. If there are metrics records, return the date of the first day without a complete metrics record.
+        2. If there are no metrics records, return the date of the first recorded session.
+        3. If there are no metrics records and no sessions records, return None.
+
+        Args:
+            table (Table): The table to get the starting date for.
+
+        Returns:
+            Optional[date]: The starting date for which metrics calculation is needed.
+        """
+        with self.Session() as sess:
+            stmt = select(table).order_by(table.c.date.desc()).limit(1)
+            result = sess.execute(stmt).fetchone()
+
+            # 1. Return the date of the first day without a complete metrics record.
+            if result is not None:
+                if result.completed:
+                    return result._mapping["date"] + timedelta(days=1)
+                else:
+                    return result._mapping["date"]
+
+        # 2. No metrics records. Return the date of the first recorded session.
+        first_session, _ = self.get_sessions(sort_by="created_at", sort_order="asc", limit=1, deserialize=False)
+        if not isinstance(first_session, list):
+            raise ValueError("Error obtaining session list to calculate metrics")
+
+        first_session_date = first_session[0]["created_at"] if first_session else None
+
+        # 3. No metrics records and no sessions records. Return None.
+        if first_session_date is None:
+            return None
+
+        return datetime.fromtimestamp(first_session_date, tz=timezone.utc).date()
+
+    def calculate_metrics(self) -> Optional[list[dict]]:
+        """Calculate metrics for all dates without complete metrics.
+
+        Returns:
+            Optional[list[dict]]: The calculated metrics.
+
+        Raises:
+            Exception: If an error occurs during metrics calculation.
+        """
+        try:
+            table = self._get_table(table_type="metrics", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            starting_date = self._get_metrics_calculation_starting_date(table)
+            if starting_date is None:
+                log_info("No session data found. Won't calculate metrics.")
+                return None
+
+            dates_to_process = get_dates_to_calculate_metrics_for(starting_date)
+            if not dates_to_process:
+                log_info("Metrics already calculated for all relevant dates.")
+                return None
+
+            start_timestamp = int(
+                datetime.combine(dates_to_process[0], datetime.min.time()).replace(tzinfo=timezone.utc).timestamp()
+            )
+            end_timestamp = int(
+                datetime.combine(dates_to_process[-1] + timedelta(days=1), datetime.min.time())
+                .replace(tzinfo=timezone.utc)
+                .timestamp()
+            )
+
+            sessions = self._get_all_sessions_for_metrics_calculation(
+                start_timestamp=start_timestamp, end_timestamp=end_timestamp
+            )
+            all_sessions_data = fetch_all_sessions_data(
+                sessions=sessions, dates_to_process=dates_to_process, start_timestamp=start_timestamp
+            )
+            if not all_sessions_data:
+                log_info("No new session data found. Won't calculate metrics.")
+                return None
+
+            results = []
+            metrics_records = []
+
+            for date_to_process in dates_to_process:
+                date_key = date_to_process.isoformat()
+                sessions_for_date = all_sessions_data.get(date_key, {})
+
+                # Skip dates with no sessions
+                if not any(len(sessions) > 0 for sessions in sessions_for_date.values()):
+                    continue
+
+                metrics_record = calculate_date_metrics(date_to_process, sessions_for_date)
+                metrics_records.append(metrics_record)
+
+            if metrics_records:
+                with self.Session() as sess, sess.begin():
+                    results = bulk_upsert_metrics(session=sess, table=table, metrics_records=metrics_records)
+
+            return results
+
+        except Exception as e:
+            log_error(f"Exception refreshing metrics: {e}")
+            return None
+
+    def get_metrics(
+        self,
+        starting_date: Optional[date] = None,
+        ending_date: Optional[date] = None,
+    ) -> Tuple[List[dict], Optional[int]]:
+        """Get all metrics matching the given date range.
+
+        Args:
+            starting_date (Optional[date]): The starting date to filter metrics by.
+            ending_date (Optional[date]): The ending date to filter metrics by.
+
+        Returns:
+            Tuple[List[dict], Optional[int]]: A tuple containing the metrics and the timestamp of the latest update.
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="metrics", create_table_if_not_found=True)
+            if table is None:
+                return [], 0
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+                if starting_date:
+                    stmt = stmt.where(table.c.date >= starting_date)
+                if ending_date:
+                    stmt = stmt.where(table.c.date <= ending_date)
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [], None
+
+                # Get the latest updated_at
+                latest_stmt = select(func.max(table.c.updated_at))
+                latest_updated_at = sess.execute(latest_stmt).scalar()
+
+            return [row._mapping for row in result], latest_updated_at
+
+        except Exception as e:
+            log_error(f"Exception getting metrics: {e}")
+            return [], None
+
+    # -- Knowledge methods --
+
+    def delete_knowledge_content(self, id: str):
+        """Delete a knowledge row from the database.
+
+        Args:
+            id (str): The ID of the knowledge row to delete.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        table = self._get_table(table_type="knowledge")
+        if table is None:
+            return None
+
+        try:
+            with self.Session() as sess, sess.begin():
+                stmt = table.delete().where(table.c.id == id)
+                sess.execute(stmt)
+
+        except Exception as e:
+            log_error(f"Exception deleting knowledge content: {e}")
+
+    def get_knowledge_content(self, id: str) -> Optional[KnowledgeRow]:
+        """Get a knowledge row from the database.
+
+        Args:
+            id (str): The ID of the knowledge row to get.
+
+        Returns:
+            Optional[KnowledgeRow]: The knowledge row, or None if it doesn't exist.
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        table = self._get_table(table_type="knowledge")
+        if table is None:
+            return None
+
+        try:
+            with self.Session() as sess, sess.begin():
+                stmt = select(table).where(table.c.id == id)
+                result = sess.execute(stmt).fetchone()
+                if result is None:
+                    return None
+                return KnowledgeRow.model_validate(result._mapping)
+
+        except Exception as e:
+            log_error(f"Exception getting knowledge content: {e}")
+            return None
+
+    def get_knowledge_contents(
+        self,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+    ) -> Tuple[List[KnowledgeRow], int]:
+        """Get all knowledge contents from the database.
+
+        Args:
+            limit (Optional[int]): The maximum number of knowledge contents to return.
+            page (Optional[int]): The page number.
+            sort_by (Optional[str]): The column to sort by.
+            sort_order (Optional[str]): The order to sort by.
+            create_table_if_not_found (Optional[bool]): Whether to create the table if it doesn't exist.
+
+        Returns:
+            Tuple[List[KnowledgeRow], int]: The knowledge contents and total count.
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        table = self._get_table(table_type="knowledge")
+        if table is None:
+            return [], 0
+
+        try:
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+
+                # Apply sorting
+                if sort_by is not None:
+                    stmt = stmt.order_by(getattr(table.c, sort_by) * (1 if sort_order == "asc" else -1))
+
+                # Get total count before applying limit and pagination
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Apply pagination after count
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [], 0
+
+                return [KnowledgeRow.model_validate(record._mapping) for record in result], total_count
+
+        except Exception as e:
+            log_error(f"Exception getting knowledge contents: {e}")
+            return [], 0
+
+    def upsert_knowledge_content(self, knowledge_row: KnowledgeRow):
+        """Upsert knowledge content in the database.
+
+        Args:
+            knowledge_row (KnowledgeRow): The knowledge row to upsert.
+
+        Returns:
+            Optional[KnowledgeRow]: The upserted knowledge row, or None if the operation fails.
+
+        Raises:
+            Exception: If an error occurs during upsert.
+        """
+        try:
+            table = self._get_table(table_type="knowledge", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                # Get the actual table columns to avoid "unconsumed column names" error
+                table_columns = set(table.columns.keys())
+
+                # Only include fields that exist in the table and are not None
+                insert_data = {}
+                update_fields = {}
+
+                # Map of KnowledgeRow fields to table columns
+                field_mapping = {
+                    "id": "id",
+                    "name": "name",
+                    "description": "description",
+                    "metadata": "metadata",
+                    "type": "type",
+                    "size": "size",
+                    "linked_to": "linked_to",
+                    "access_count": "access_count",
+                    "status": "status",
+                    "status_message": "status_message",
+                    "created_at": "created_at",
+                    "updated_at": "updated_at",
+                    "external_id": "external_id",
+                }
+
+                # Build insert and update data only for fields that exist in the table
+                for model_field, table_column in field_mapping.items():
+                    if table_column in table_columns:
+                        value = getattr(knowledge_row, model_field, None)
+                        if value is not None:
+                            insert_data[table_column] = value
+                            # Don't include ID in update_fields since it's the primary key
+                            if table_column != "id":
+                                update_fields[table_column] = value
+
+                # Ensure id is always included for the insert
+                if "id" in table_columns and knowledge_row.id:
+                    insert_data["id"] = knowledge_row.id
+
+                # Handle case where update_fields is empty (all fields are None or don't exist in table)
+                if not update_fields:
+                    # If we have insert_data, just do an insert without conflict resolution
+                    if insert_data:
+                        stmt = mysql.insert(table).values(insert_data)
+                        sess.execute(stmt)
+                    else:
+                        # If we have no data at all, this is an error
+                        log_error("No valid fields found for knowledge row upsert")
+                        return None
+                else:
+                    # Normal upsert with conflict resolution
+                    stmt = mysql.insert(table).values(insert_data).on_duplicate_key_update(**update_fields)
+                    sess.execute(stmt)
+
+            return knowledge_row
+
+        except Exception as e:
+            log_error(f"Error upserting knowledge row: {e}")
+            return None
+
+    # -- Eval methods --
+
+    def create_eval_run(self, eval_run: EvalRunRecord) -> Optional[EvalRunRecord]:
+        """Create an EvalRunRecord in the database.
+
+        Args:
+            eval_run (EvalRunRecord): The eval run to create.
+
+        Returns:
+            Optional[EvalRunRecord]: The created eval run, or None if the operation fails.
+
+        Raises:
+            Exception: If an error occurs during creation.
+        """
+        try:
+            table = self._get_table(table_type="evals", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                current_time = int(time.time())
+                stmt = mysql.insert(table).values(
+                    {"created_at": current_time, "updated_at": current_time, **eval_run.model_dump()}
+                )
+                sess.execute(stmt)
+
+            return eval_run
+
+        except Exception as e:
+            log_error(f"Error creating eval run: {e}")
+            return None
+
+    def delete_eval_run(self, eval_run_id: str) -> None:
+        """Delete an eval run from the database.
+
+        Args:
+            eval_run_id (str): The ID of the eval run to delete.
+        """
+        try:
+            table = self._get_table(table_type="evals")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                stmt = table.delete().where(table.c.run_id == eval_run_id)
+                result = sess.execute(stmt)
+                if result.rowcount == 0:
+                    log_error(f"No eval run found with ID: {eval_run_id}")
+                else:
+                    log_debug(f"Deleted eval run with ID: {eval_run_id}")
+
+        except Exception as e:
+            log_error(f"Error deleting eval run {eval_run_id}: {e}")
+
+    def delete_eval_runs(self, eval_run_ids: List[str]) -> None:
+        """Delete multiple eval runs from the database.
+
+        Args:
+            eval_run_ids (List[str]): List of eval run IDs to delete.
+        """
+        try:
+            table = self._get_table(table_type="evals")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                stmt = table.delete().where(table.c.run_id.in_(eval_run_ids))
+                result = sess.execute(stmt)
+                if result.rowcount == 0:
+                    log_error(f"No eval runs found with IDs: {eval_run_ids}")
+                else:
+                    log_debug(f"Deleted {result.rowcount} eval runs")
+
+        except Exception as e:
+            log_error(f"Error deleting eval runs {eval_run_ids}: {e}")
+
+    def get_eval_run(
+        self, eval_run_id: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[EvalRunRecord, Dict[str, Any]]]:
+        """Get an eval run from the database.
+
+        Args:
+            eval_run_id (str): The ID of the eval run to get.
+            deserialize (Optional[bool]): Whether to serialize the eval run. Defaults to True.
+
+        Returns:
+            Optional[Union[EvalRunRecord, Dict[str, Any]]]:
+                - When deserialize=True: EvalRunRecord object
+                - When deserialize=False: EvalRun dictionary
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="evals")
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table).where(table.c.run_id == eval_run_id)
+                result = sess.execute(stmt).fetchone()
+                if result is None:
+                    return None
+
+                eval_run_raw = result._mapping
+                if not deserialize:
+                    return eval_run_raw
+
+                return EvalRunRecord.model_validate(eval_run_raw)
+
+        except Exception as e:
+            log_error(f"Exception getting eval run {eval_run_id}: {e}")
+            return None
+
+    def get_eval_runs(
+        self,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        workflow_id: Optional[str] = None,
+        model_id: Optional[str] = None,
+        filter_type: Optional[EvalFilterType] = None,
+        eval_type: Optional[List[EvalType]] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[EvalRunRecord], Tuple[List[Dict[str, Any]], int]]:
+        """Get all eval runs from the database.
+
+        Args:
+            limit (Optional[int]): The maximum number of eval runs to return.
+            page (Optional[int]): The page number.
+            sort_by (Optional[str]): The column to sort by.
+            sort_order (Optional[str]): The order to sort by.
+            agent_id (Optional[str]): The ID of the agent to filter by.
+            team_id (Optional[str]): The ID of the team to filter by.
+            workflow_id (Optional[str]): The ID of the workflow to filter by.
+            model_id (Optional[str]): The ID of the model to filter by.
+            eval_type (Optional[List[EvalType]]): The type(s) of eval to filter by.
+            filter_type (Optional[EvalFilterType]): Filter by component type (agent, team, workflow).
+            deserialize (Optional[bool]): Whether to serialize the eval runs. Defaults to True.
+            create_table_if_not_found (Optional[bool]): Whether to create the table if it doesn't exist.
+
+        Returns:
+            Union[List[EvalRunRecord], Tuple[List[Dict[str, Any]], int]]:
+                - When deserialize=True: List of EvalRunRecord objects
+                - When deserialize=False: List of dictionaries
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="evals")
+            if table is None:
+                return [] if deserialize else ([], 0)
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+
+                # Filtering
+                if agent_id is not None:
+                    stmt = stmt.where(table.c.agent_id == agent_id)
+                if team_id is not None:
+                    stmt = stmt.where(table.c.team_id == team_id)
+                if workflow_id is not None:
+                    stmt = stmt.where(table.c.workflow_id == workflow_id)
+                if model_id is not None:
+                    stmt = stmt.where(table.c.model_id == model_id)
+                if eval_type is not None and len(eval_type) > 0:
+                    stmt = stmt.where(table.c.eval_type.in_(eval_type))
+                if filter_type is not None:
+                    if filter_type == EvalFilterType.AGENT:
+                        stmt = stmt.where(table.c.agent_id.is_not(None))
+                    elif filter_type == EvalFilterType.TEAM:
+                        stmt = stmt.where(table.c.team_id.is_not(None))
+                    elif filter_type == EvalFilterType.WORKFLOW:
+                        stmt = stmt.where(table.c.workflow_id.is_not(None))
+
+                # Get total count after applying filtering
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Sorting
+                if sort_by is None:
+                    stmt = stmt.order_by(table.c.created_at.desc())
+                else:
+                    stmt = apply_sorting(stmt, table, sort_by, sort_order)
+
+                # Paginating
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [] if deserialize else ([], 0)
+
+                eval_runs_raw = [row._mapping for row in result]
+                if not deserialize:
+                    return eval_runs_raw, total_count
+
+                return [EvalRunRecord.model_validate(row) for row in eval_runs_raw]
+
+        except Exception as e:
+            log_error(f"Exception getting eval runs: {e}")
+            raise e
+
+    def rename_eval_run(
+        self, eval_run_id: str, name: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[EvalRunRecord, Dict[str, Any]]]:
+        """Upsert the name of an eval run in the database, returning raw dictionary.
+
+        Args:
+            eval_run_id (str): The ID of the eval run to update.
+            name (str): The new name of the eval run.
+
+        Returns:
+            Optional[Dict[str, Any]]: The updated eval run, or None if the operation fails.
+
+        Raises:
+            Exception: If an error occurs during update.
+        """
+        try:
+            table = self._get_table(table_type="evals")
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                stmt = (
+                    table.update().where(table.c.run_id == eval_run_id).values(name=name, updated_at=int(time.time()))
+                )
+                sess.execute(stmt)
+
+            eval_run_raw = self.get_eval_run(eval_run_id=eval_run_id, deserialize=deserialize)
+            if not eval_run_raw or not deserialize:
+                return eval_run_raw
+
+            return EvalRunRecord.model_validate(eval_run_raw)
+
+        except Exception as e:
+            log_error(f"Error upserting eval run name {eval_run_id}: {e}")
+            return None
+
+    # -- Culture methods --
+
+    def clear_cultural_knowledge(self) -> None:
+        """Delete all cultural knowledge from the database.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="culture")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                sess.execute(table.delete())
+
+        except Exception as e:
+            log_warning(f"Exception deleting all cultural knowledge: {e}")
+            raise e
+
+    def delete_cultural_knowledge(self, id: str) -> None:
+        """Delete a cultural knowledge entry from the database.
+
+        Args:
+            id (str): The ID of the cultural knowledge to delete.
+
+        Raises:
+            Exception: If an error occurs during deletion.
+        """
+        try:
+            table = self._get_table(table_type="culture")
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                delete_stmt = table.delete().where(table.c.id == id)
+                result = sess.execute(delete_stmt)
+
+                success = result.rowcount > 0
+                if success:
+                    log_debug(f"Successfully deleted cultural knowledge id: {id}")
+                else:
+                    log_debug(f"No cultural knowledge found with id: {id}")
+
+        except Exception as e:
+            log_error(f"Error deleting cultural knowledge: {e}")
+            raise e
+
+    def get_cultural_knowledge(
+        self, id: str, deserialize: Optional[bool] = True
+    ) -> Optional[Union[CulturalKnowledge, Dict[str, Any]]]:
+        """Get a cultural knowledge entry from the database.
+
+        Args:
+            id (str): The ID of the cultural knowledge to get.
+            deserialize (Optional[bool]): Whether to deserialize the cultural knowledge. Defaults to True.
+
+        Returns:
+            Optional[Union[CulturalKnowledge, Dict[str, Any]]]: The cultural knowledge entry, or None if it doesn't exist.
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="culture")
+            if table is None:
+                return None
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table).where(table.c.id == id)
+                result = sess.execute(stmt).fetchone()
+                if result is None:
+                    return None
+
+                db_row = dict(result._mapping)
+                if not db_row or not deserialize:
+                    return db_row
+
+            return deserialize_cultural_knowledge_from_db(db_row)
+
+        except Exception as e:
+            log_error(f"Exception reading from cultural knowledge table: {e}")
+            raise e
+
+    def get_all_cultural_knowledge(
+        self,
+        name: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        limit: Optional[int] = None,
+        page: Optional[int] = None,
+        sort_by: Optional[str] = None,
+        sort_order: Optional[str] = None,
+        deserialize: Optional[bool] = True,
+    ) -> Union[List[CulturalKnowledge], Tuple[List[Dict[str, Any]], int]]:
+        """Get all cultural knowledge from the database as CulturalKnowledge objects.
+
+        Args:
+            name (Optional[str]): The name of the cultural knowledge to filter by.
+            agent_id (Optional[str]): The ID of the agent to filter by.
+            team_id (Optional[str]): The ID of the team to filter by.
+            limit (Optional[int]): The maximum number of cultural knowledge entries to return.
+            page (Optional[int]): The page number.
+            sort_by (Optional[str]): The column to sort by.
+            sort_order (Optional[str]): The order to sort by.
+            deserialize (Optional[bool]): Whether to deserialize the cultural knowledge. Defaults to True.
+
+        Returns:
+            Union[List[CulturalKnowledge], Tuple[List[Dict[str, Any]], int]]:
+                - When deserialize=True: List of CulturalKnowledge objects
+                - When deserialize=False: List of CulturalKnowledge dictionaries and total count
+
+        Raises:
+            Exception: If an error occurs during retrieval.
+        """
+        try:
+            table = self._get_table(table_type="culture")
+            if table is None:
+                return [] if deserialize else ([], 0)
+
+            with self.Session() as sess, sess.begin():
+                stmt = select(table)
+
+                # Filtering
+                if name is not None:
+                    stmt = stmt.where(table.c.name == name)
+                if agent_id is not None:
+                    stmt = stmt.where(table.c.agent_id == agent_id)
+                if team_id is not None:
+                    stmt = stmt.where(table.c.team_id == team_id)
+
+                # Get total count after applying filtering
+                count_stmt = select(func.count()).select_from(stmt.alias())
+                total_count = sess.execute(count_stmt).scalar()
+
+                # Sorting
+                stmt = apply_sorting(stmt, table, sort_by, sort_order)
+                # Paginating
+                if limit is not None:
+                    stmt = stmt.limit(limit)
+                    if page is not None:
+                        stmt = stmt.offset((page - 1) * limit)
+
+                result = sess.execute(stmt).fetchall()
+                if not result:
+                    return [] if deserialize else ([], 0)
+
+                db_rows = [dict(record._mapping) for record in result]
+
+                if not deserialize:
+                    return db_rows, total_count
+
+            return [deserialize_cultural_knowledge_from_db(row) for row in db_rows]
+
+        except Exception as e:
+            log_error(f"Error reading from cultural knowledge table: {e}")
+            raise e
+
+    def upsert_cultural_knowledge(
+        self, cultural_knowledge: CulturalKnowledge, deserialize: Optional[bool] = True
+    ) -> Optional[Union[CulturalKnowledge, Dict[str, Any]]]:
+        """Upsert a cultural knowledge entry into the database.
+
+        Args:
+            cultural_knowledge (CulturalKnowledge): The cultural knowledge to upsert.
+            deserialize (Optional[bool]): Whether to deserialize the cultural knowledge. Defaults to True.
+
+        Returns:
+            Optional[CulturalKnowledge]: The upserted cultural knowledge entry.
+
+        Raises:
+            Exception: If an error occurs during upsert.
+        """
+        try:
+            table = self._get_table(table_type="culture", create_table_if_not_found=True)
+            if table is None:
+                return None
+
+            if cultural_knowledge.id is None:
+                cultural_knowledge.id = str(uuid4())
+
+            # Serialize content, categories, and notes into a JSON dict for DB storage
+            content_dict = serialize_cultural_knowledge_for_db(cultural_knowledge)
+
+            with self.Session() as sess, sess.begin():
+                stmt = mysql.insert(table).values(
+                    id=cultural_knowledge.id,
+                    name=cultural_knowledge.name,
+                    summary=cultural_knowledge.summary,
+                    content=content_dict if content_dict else None,
+                    metadata=cultural_knowledge.metadata,
+                    input=cultural_knowledge.input,
+                    created_at=cultural_knowledge.created_at,
+                    updated_at=int(time.time()),
+                    agent_id=cultural_knowledge.agent_id,
+                    team_id=cultural_knowledge.team_id,
+                )
+                stmt = stmt.on_duplicate_key_update(
+                    name=cultural_knowledge.name,
+                    summary=cultural_knowledge.summary,
+                    content=content_dict if content_dict else None,
+                    metadata=cultural_knowledge.metadata,
+                    input=cultural_knowledge.input,
+                    updated_at=int(time.time()),
+                    agent_id=cultural_knowledge.agent_id,
+                    team_id=cultural_knowledge.team_id,
+                )
+                sess.execute(stmt)
+
+            # Fetch the inserted/updated row
+            return self.get_cultural_knowledge(id=cultural_knowledge.id, deserialize=deserialize)
+
+        except Exception as e:
+            log_error(f"Error upserting cultural knowledge: {e}")
+            raise e
+
+    # -- Migrations --
+
+    def migrate_table_from_v1_to_v2(self, v1_db_schema: str, v1_table_name: str, v1_table_type: str):
+        """Migrate all content in the given table to the right v2 table"""
+
+        from agno.db.migrations.v1_to_v2 import (
+            get_all_table_content,
+            parse_agent_sessions,
+            parse_memories,
+            parse_team_sessions,
+            parse_workflow_sessions,
+        )
+
+        # Get all content from the old table
+        old_content: list[dict[str, Any]] = get_all_table_content(
+            db=self,
+            db_schema=v1_db_schema,
+            table_name=v1_table_name,
+        )
+        if not old_content:
+            log_info(f"No content to migrate from table {v1_table_name}")
+            return
+
+        # Parse the content into the new format
+        memories: List[UserMemory] = []
+        sessions: Sequence[Union[AgentSession, TeamSession, WorkflowSession]] = []
+        if v1_table_type == "agent_sessions":
+            sessions = parse_agent_sessions(old_content)
+        elif v1_table_type == "team_sessions":
+            sessions = parse_team_sessions(old_content)
+        elif v1_table_type == "workflow_sessions":
+            sessions = parse_workflow_sessions(old_content)
+        elif v1_table_type == "memories":
+            memories = parse_memories(old_content)
+        else:
+            raise ValueError(f"Invalid table type: {v1_table_type}")
+
+        # Insert the new content into the new table
+        if v1_table_type == "agent_sessions":
+            for session in sessions:
+                self.upsert_session(session)
+            log_info(f"Migrated {len(sessions)} Agent sessions to table: {self.session_table_name}")
+
+        elif v1_table_type == "team_sessions":
+            for session in sessions:
+                self.upsert_session(session)
+            log_info(f"Migrated {len(sessions)} Team sessions to table: {self.session_table_name}")
+
+        elif v1_table_type == "workflow_sessions":
+            for session in sessions:
+                self.upsert_session(session)
+            log_info(f"Migrated {len(sessions)} Workflow sessions to table: {self.session_table_name}")
+
+        elif v1_table_type == "memories":
+            for memory in memories:
+                self.upsert_user_memory(memory)
+            log_info(f"Migrated {len(memories)} memories to table: {self.memory_table}")
+
+    # --- Traces ---
+    def _get_traces_base_query(self, table: Table, spans_table: Optional[Table] = None):
+        """Build base query for traces with aggregated span counts.
+
+        Args:
+            table: The traces table.
+            spans_table: The spans table (optional).
+
+        Returns:
+            SQLAlchemy select statement with total_spans and error_count calculated dynamically.
+        """
+        from sqlalchemy import case, literal
+
+        if spans_table is not None:
+            # JOIN with spans table to calculate total_spans and error_count
+            return (
+                select(
+                    table,
+                    func.coalesce(func.count(spans_table.c.span_id), 0).label("total_spans"),
+                    func.coalesce(func.sum(case((spans_table.c.status_code == "ERROR", 1), else_=0)), 0).label(
+                        "error_count"
+                    ),
+                )
+                .select_from(table.outerjoin(spans_table, table.c.trace_id == spans_table.c.trace_id))
+                .group_by(table.c.trace_id)
+            )
+        else:
+            # Fallback if spans table doesn't exist
+            return select(table, literal(0).label("total_spans"), literal(0).label("error_count"))
+
+    def _get_trace_component_level_expr(self, workflow_id_col, team_id_col, agent_id_col, name_col):
+        """Build a SQL CASE expression that returns the component level for a trace.
+
+        Component levels (higher = more important):
+            - 3: Workflow root (.run or .arun with workflow_id)
+            - 2: Team root (.run or .arun with team_id)
+            - 1: Agent root (.run or .arun with agent_id)
+            - 0: Child span (not a root)
+
+        Args:
+            workflow_id_col: SQL column/expression for workflow_id
+            team_id_col: SQL column/expression for team_id
+            agent_id_col: SQL column/expression for agent_id
+            name_col: SQL column/expression for name
+
+        Returns:
+            SQLAlchemy CASE expression returning the component level as an integer.
+        """
+        from sqlalchemy import and_, case, or_
+
+        is_root_name = or_(name_col.like("%.run%"), name_col.like("%.arun%"))
+
+        return case(
+            # Workflow root (level 3)
+            (and_(workflow_id_col.isnot(None), is_root_name), 3),
+            # Team root (level 2)
+            (and_(team_id_col.isnot(None), is_root_name), 2),
+            # Agent root (level 1)
+            (and_(agent_id_col.isnot(None), is_root_name), 1),
+            # Child span or unknown (level 0)
+            else_=0,
+        )
+
+    def upsert_trace(self, trace: "Trace") -> None:
+        """Create or update a single trace record in the database.
+
+        Uses INSERT ... ON DUPLICATE KEY UPDATE (upsert) to handle concurrent inserts
+        atomically and avoid race conditions.
+
+        Args:
+            trace: The Trace object to store (one per trace_id).
+        """
+        from sqlalchemy import case
+
+        try:
+            table = self._get_table(table_type="traces", create_table_if_not_found=True)
+            if table is None:
+                return
+
+            trace_dict = trace.to_dict()
+            trace_dict.pop("total_spans", None)
+            trace_dict.pop("error_count", None)
+
+            with self.Session() as sess, sess.begin():
+                # Use upsert to handle concurrent inserts atomically
+                # On conflict, update fields while preserving existing non-null context values
+                # and keeping the earliest start_time
+                insert_stmt = mysql.insert(table).values(trace_dict)
+
+                # Build component level expressions for comparing trace priority
+                new_level = self._get_trace_component_level_expr(
+                    insert_stmt.inserted.workflow_id,
+                    insert_stmt.inserted.team_id,
+                    insert_stmt.inserted.agent_id,
+                    insert_stmt.inserted.name,
+                )
+                existing_level = self._get_trace_component_level_expr(
+                    table.c.workflow_id,
+                    table.c.team_id,
+                    table.c.agent_id,
+                    table.c.name,
+                )
+
+                # Build the ON DUPLICATE KEY UPDATE clause
+                # Use LEAST for start_time, GREATEST for end_time to capture full trace duration
+                # MySQL stores timestamps as ISO strings, so string comparison works for ISO format
+                # Duration is calculated using TIMESTAMPDIFF in microseconds then converted to ms
+                upsert_stmt = insert_stmt.on_duplicate_key_update(
+                    end_time=func.greatest(table.c.end_time, insert_stmt.inserted.end_time),
+                    start_time=func.least(table.c.start_time, insert_stmt.inserted.start_time),
+                    # Calculate duration in milliseconds using TIMESTAMPDIFF
+                    # TIMESTAMPDIFF(MICROSECOND, start, end) / 1000 gives milliseconds
+                    duration_ms=func.timestampdiff(
+                        text("MICROSECOND"),
+                        func.least(table.c.start_time, insert_stmt.inserted.start_time),
+                        func.greatest(table.c.end_time, insert_stmt.inserted.end_time),
+                    )
+                    / 1000,
+                    status=insert_stmt.inserted.status,
+                    # Update name only if new trace is from a higher-level component
+                    # Priority: workflow (3) > team (2) > agent (1) > child spans (0)
+                    name=case(
+                        (new_level > existing_level, insert_stmt.inserted.name),
+                        else_=table.c.name,
+                    ),
+                    # Preserve existing non-null context values using COALESCE
+                    run_id=func.coalesce(insert_stmt.inserted.run_id, table.c.run_id),
+                    session_id=func.coalesce(insert_stmt.inserted.session_id, table.c.session_id),
+                    user_id=func.coalesce(insert_stmt.inserted.user_id, table.c.user_id),
+                    agent_id=func.coalesce(insert_stmt.inserted.agent_id, table.c.agent_id),
+                    team_id=func.coalesce(insert_stmt.inserted.team_id, table.c.team_id),
+                    workflow_id=func.coalesce(insert_stmt.inserted.workflow_id, table.c.workflow_id),
+                )
+                sess.execute(upsert_stmt)
+
+        except Exception as e:
+            log_error(f"Error creating trace: {e}")
+            # Don't raise - tracing should not break the main application flow
+
+    def get_trace(
+        self,
+        trace_id: Optional[str] = None,
+        run_id: Optional[str] = None,
+    ):
+        """Get a single trace by trace_id or other filters.
+
+        Args:
+            trace_id: The unique trace identifier.
+            run_id: Filter by run ID (returns first match).
+
+        Returns:
+            Optional[Trace]: The trace if found, None otherwise.
+
+        Note:
+            If multiple filters are provided, trace_id takes precedence.
+            For other filters, the most recent trace is returned.
+        """
+        try:
+            from agno.tracing.schemas import Trace
+
+            table = self._get_table(table_type="traces")
+            if table is None:
+                return None
+
+            # Get spans table for JOIN
+            spans_table = self._get_table(table_type="spans")
+
+            with self.Session() as sess:
+                # Build query with aggregated span counts
+                stmt = self._get_traces_base_query(table, spans_table)
+
+                if trace_id:
+                    stmt = stmt.where(table.c.trace_id == trace_id)
+                elif run_id:
+                    stmt = stmt.where(table.c.run_id == run_id)
+                else:
+                    log_debug("get_trace called without any filter parameters")
+                    return None
+
+                # Order by most recent and get first result
+                stmt = stmt.order_by(table.c.start_time.desc()).limit(1)
+                result = sess.execute(stmt).fetchone()
+
+                if result:
+                    return Trace.from_dict(dict(result._mapping))
+                return None
+
+        except Exception as e:
+            log_error(f"Error getting trace: {e}")
+            return None
+
+    def get_traces(
+        self,
+        run_id: Optional[str] = None,
+        session_id: Optional[str] = None,
+        user_id: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        workflow_id: Optional[str] = None,
+        status: Optional[str] = None,
+        start_time: Optional[datetime] = None,
+        end_time: Optional[datetime] = None,
+        limit: Optional[int] = 20,
+        page: Optional[int] = 1,
+    ) -> tuple[List, int]:
+        """Get traces matching the provided filters with pagination.
+
+        Args:
+            run_id: Filter by run ID.
+            session_id: Filter by session ID.
+            user_id: Filter by user ID.
+            agent_id: Filter by agent ID.
+            team_id: Filter by team ID.
+            workflow_id: Filter by workflow ID.
+            status: Filter by status (OK, ERROR, UNSET).
+            start_time: Filter traces starting after this datetime.
+            end_time: Filter traces ending before this datetime.
+            limit: Maximum number of traces to return per page.
+            page: Page number (1-indexed).
+
+        Returns:
+            tuple[List[Trace], int]: Tuple of (list of matching traces, total count).
+        """
+        try:
+            from agno.tracing.schemas import Trace
+
+            log_debug(
+                f"get_traces called with filters: run_id={run_id}, session_id={session_id}, user_id={user_id}, agent_id={agent_id}, page={page}, limit={limit}"
+            )
+
+            table = self._get_table(table_type="traces")
+            if table is None:
+                log_debug("Traces table not found")
+                return [], 0
+
+            # Get spans table for JOIN
+            spans_table = self._get_table(table_type="spans")
+
+            with self.Session() as sess:
+                # Build base query with aggregated span counts
+                base_stmt = self._get_traces_base_query(table, spans_table)
+
+                # Apply filters
+                if run_id:
+                    base_stmt = base_stmt.where(table.c.run_id == run_id)
+                if session_id:
+                    base_stmt = base_stmt.where(table.c.session_id == session_id)
+                if user_id:
+                    base_stmt = base_stmt.where(table.c.user_id == user_id)
+                if agent_id:
+                    base_stmt = base_stmt.where(table.c.agent_id == agent_id)
+                if team_id:
+                    base_stmt = base_stmt.where(table.c.team_id == team_id)
+                if workflow_id:
+                    base_stmt = base_stmt.where(table.c.workflow_id == workflow_id)
+                if status:
+                    base_stmt = base_stmt.where(table.c.status == status)
+                if start_time:
+                    # Convert datetime to ISO string for comparison
+                    base_stmt = base_stmt.where(table.c.start_time >= start_time.isoformat())
+                if end_time:
+                    # Convert datetime to ISO string for comparison
+                    base_stmt = base_stmt.where(table.c.end_time <= end_time.isoformat())
+
+                # Get total count
+                count_stmt = select(func.count()).select_from(base_stmt.alias())
+                total_count = sess.execute(count_stmt).scalar() or 0
+
+                # Apply pagination
+                offset = (page - 1) * limit if page and limit else 0
+                paginated_stmt = base_stmt.order_by(table.c.start_time.desc()).limit(limit).offset(offset)
+
+                results = sess.execute(paginated_stmt).fetchall()
+
+                traces = [Trace.from_dict(dict(row._mapping)) for row in results]
+                return traces, total_count
+
+        except Exception as e:
+            log_error(f"Error getting traces: {e}")
+            return [], 0
+
+    def get_trace_stats(
+        self,
+        user_id: Optional[str] = None,
+        agent_id: Optional[str] = None,
+        team_id: Optional[str] = None,
+        workflow_id: Optional[str] = None,
+        start_time: Optional[datetime] = None,
+        end_time: Optional[datetime] = None,
+        limit: Optional[int] = 20,
+        page: Optional[int] = 1,
+    ) -> tuple[List[Dict[str, Any]], int]:
+        """Get trace statistics grouped by session.
+
+        Args:
+            user_id: Filter by user ID.
+            agent_id: Filter by agent ID.
+            team_id: Filter by team ID.
+            workflow_id: Filter by workflow ID.
+            start_time: Filter sessions with traces created after this datetime.
+            end_time: Filter sessions with traces created before this datetime.
+            limit: Maximum number of sessions to return per page.
+            page: Page number (1-indexed).
+
+        Returns:
+            tuple[List[Dict], int]: Tuple of (list of session stats dicts, total count).
+                Each dict contains: session_id, user_id, agent_id, team_id, total_traces,
+                workflow_id, first_trace_at, last_trace_at.
+        """
+        try:
+            table = self._get_table(table_type="traces")
+            if table is None:
+                log_debug("Traces table not found")
+                return [], 0
+
+            with self.Session() as sess:
+                # Build base query grouped by session_id
+                base_stmt = (
+                    select(
+                        table.c.session_id,
+                        table.c.user_id,
+                        table.c.agent_id,
+                        table.c.team_id,
+                        table.c.workflow_id,
+                        func.count(table.c.trace_id).label("total_traces"),
+                        func.min(table.c.created_at).label("first_trace_at"),
+                        func.max(table.c.created_at).label("last_trace_at"),
+                    )
+                    .where(table.c.session_id.isnot(None))  # Only sessions with session_id
+                    .group_by(
+                        table.c.session_id, table.c.user_id, table.c.agent_id, table.c.team_id, table.c.workflow_id
+                    )
+                )
+
+                # Apply filters
+                if user_id:
+                    base_stmt = base_stmt.where(table.c.user_id == user_id)
+                if workflow_id:
+                    base_stmt = base_stmt.where(table.c.workflow_id == workflow_id)
+                if team_id:
+                    base_stmt = base_stmt.where(table.c.team_id == team_id)
+                if agent_id:
+                    base_stmt = base_stmt.where(table.c.agent_id == agent_id)
+                if start_time:
+                    # Convert datetime to ISO string for comparison
+                    base_stmt = base_stmt.where(table.c.created_at >= start_time.isoformat())
+                if end_time:
+                    # Convert datetime to ISO string for comparison
+                    base_stmt = base_stmt.where(table.c.created_at <= end_time.isoformat())
+
+                # Get total count of sessions
+                count_stmt = select(func.count()).select_from(base_stmt.alias())
+                total_count = sess.execute(count_stmt).scalar() or 0
+
+                # Apply pagination and ordering
+                offset = (page - 1) * limit if page and limit else 0
+                paginated_stmt = base_stmt.order_by(func.max(table.c.created_at).desc()).limit(limit).offset(offset)
+
+                results = sess.execute(paginated_stmt).fetchall()
+
+                # Convert to list of dicts with datetime objects
+                stats_list = []
+                for row in results:
+                    # Convert ISO strings to datetime objects
+                    first_trace_at_str = row.first_trace_at
+                    last_trace_at_str = row.last_trace_at
+
+                    # Parse ISO format strings to datetime objects
+                    first_trace_at = datetime.fromisoformat(first_trace_at_str.replace("Z", "+00:00"))
+                    last_trace_at = datetime.fromisoformat(last_trace_at_str.replace("Z", "+00:00"))
+
+                    stats_list.append(
+                        {
+                            "session_id": row.session_id,
+                            "user_id": row.user_id,
+                            "agent_id": row.agent_id,
+                            "team_id": row.team_id,
+                            "workflow_id": row.workflow_id,
+                            "total_traces": row.total_traces,
+                            "first_trace_at": first_trace_at,
+                            "last_trace_at": last_trace_at,
+                        }
+                    )
+
+                return stats_list, total_count
+
+        except Exception as e:
+            log_error(f"Error getting trace stats: {e}")
+            return [], 0
+
+    # --- Spans ---
+    def create_span(self, span: "Span") -> None:
+        """Create a single span in the database.
+
+        Args:
+            span: The Span object to store.
+        """
+        try:
+            table = self._get_table(table_type="spans", create_table_if_not_found=True)
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                stmt = mysql.insert(table).values(span.to_dict())
+                sess.execute(stmt)
+
+        except Exception as e:
+            log_error(f"Error creating span: {e}")
+
+    def create_spans(self, spans: List) -> None:
+        """Create multiple spans in the database as a batch.
+
+        Args:
+            spans: List of Span objects to store.
+        """
+        if not spans:
+            return
+
+        try:
+            table = self._get_table(table_type="spans", create_table_if_not_found=True)
+            if table is None:
+                return
+
+            with self.Session() as sess, sess.begin():
+                for span in spans:
+                    stmt = mysql.insert(table).values(span.to_dict())
+                    sess.execute(stmt)
+
+        except Exception as e:
+            log_error(f"Error creating spans batch: {e}")
+
+    def get_span(self, span_id: str):
+        """Get a single span by its span_id.
+
+        Args:
+            span_id: The unique span identifier.
+
+        Returns:
+            Optional[Span]: The span if found, None otherwise.
+        """
+        try:
+            from agno.tracing.schemas import Span
+
+            table = self._get_table(table_type="spans")
+            if table is None:
+                return None
+
+            with self.Session() as sess:
+                stmt = select(table).where(table.c.span_id == span_id)
+                result = sess.execute(stmt).fetchone()
+                if result:
+                    return Span.from_dict(dict(result._mapping))
+                return None
+
+        except Exception as e:
+            log_error(f"Error getting span: {e}")
+            return None
+
+    def get_spans(
+        self,
+        trace_id: Optional[str] = None,
+        parent_span_id: Optional[str] = None,
+        limit: Optional[int] = 1000,
+    ) -> List:
+        """Get spans matching the provided filters.
+
+        Args:
+            trace_id: Filter by trace ID.
+            parent_span_id: Filter by parent span ID.
+            limit: Maximum number of spans to return.
+
+        Returns:
+            List[Span]: List of matching spans.
+        """
+        try:
+            from agno.tracing.schemas import Span
+
+            table = self._get_table(table_type="spans")
+            if table is None:
+                return []
+
+            with self.Session() as sess:
+                stmt = select(table)
+
+                # Apply filters
+                if trace_id:
+                    stmt = stmt.where(table.c.trace_id == trace_id)
+                if parent_span_id:
+                    stmt = stmt.where(table.c.parent_span_id == parent_span_id)
+
+                if limit:
+                    stmt = stmt.limit(limit)
+
+                results = sess.execute(stmt).fetchall()
+                return [Span.from_dict(dict(row._mapping)) for row in results]
+
+        except Exception as e:
+            log_error(f"Error getting spans: {e}")
+            return []