diff --git a/src/aare/common/exception_handler.py b/src/aare/common/exception_handler.py
index 72e4bce3..909a0a7f 100644
--- a/src/aare/common/exception_handler.py
+++ b/src/aare/common/exception_handler.py
@@ -1,15 +1,81 @@
 from __future__ import annotations
 
 import time
+from typing import ClassVar
 
 from aare.common.logger_config import setup_logger
 from aare.common.error_codes import AuthErrorCode, DAQErrorCode
 
 logger = setup_logger("aareDAQ")
 
-class TransformationInvalidException(Exception):
-    def __init__(self, message: str = "Transformation is not implemented"):
-        super().__init__(message)
+
+class AareException(Exception):
+    """Root of the aare exception hierarchy.
+
+    Class-level ``critical`` is the default; pass ``critical=...`` to the
+    constructor of an "optionally critical" subclass to override on a single
+    raise site.
+    """
+
+    critical: ClassVar[bool] = False
+
+    def __init__(self, *args, critical: bool | None = None, **kwargs):
+        super().__init__(*args, **kwargs)
+        if critical is not None:
+            self.critical = critical
+
+
+class AutomationError(AareException):
+    """Errors that happen during a DAQ operation; route through automation flows."""
+
+
+class AareUserError(AareException):
+    """Bad input / mode misuse; routes through the input-correction flow."""
+
+
+class AareAuthError(AareException):
+    """Authentication / authorization failures; routes through re-auth flow."""
+
+
+# ---------------------------------------------------------------------------
+# Device-family base classes
+#
+# These exist purely so watchers can match a whole device family via
+# ``isinstance(e, TellException)`` etc. Per the plan they are intentionally
+# empty -- do not add behavior.
+# ---------------------------------------------------------------------------
+
+class TellException(AutomationError):
+    pass
+
+
+class SmargonException(AutomationError):
+    pass
+
+
+class AerotechException(AutomationError):
+    pass
+
+
+class JFJochException(AutomationError):
+    pass
+
+
+class BECException(AutomationError):
+    pass
+
+
+class BeamlineStateException(AutomationError):
+    pass
+
+
+# ---------------------------------------------------------------------------
+# Concrete automation exceptions
+# ---------------------------------------------------------------------------
+
+class TransformationInvalidException(AutomationError):
+    def __init__(self, message: str = "Transformation is not implemented", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -17,9 +83,9 @@ class TransformationInvalidException(Exception):
         return self.message
 
 
-class StateTransitionFailed(Exception):
-    def __init__(self, message: str = "Beamline state transition failed"):
-        super().__init__(message)
+class StateTransitionFailed(BeamlineStateException):
+    def __init__(self, message: str = "Beamline state transition failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -27,9 +93,9 @@ class StateTransitionFailed(Exception):
         return self.message
 
 
-class MaintenanceStateException(Exception):
-    def __init__(self, message: str = "Beamline is in Maintenance state"):
-        super().__init__(message)
+class MaintenanceStateException(BeamlineStateException):
+    def __init__(self, message: str = "Beamline is in Maintenance state", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -37,31 +103,33 @@ class MaintenanceStateException(Exception):
         return self.message
 
 
-class DataCollectionException(Exception):
-    def __init__(self, message: str = "Data collection failed"):
-        super().__init__(message)
+class DataCollectionException(AutomationError):
+    """Group parent for data-collection-time exceptions; also raisable directly."""
+
+    def __init__(self, message: str = "Data collection failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
 
     def __str__(self) -> str:
         return self.message
 
 
-class RasterScanException(Exception):
-    def __init__(self, message: str = "Data collection failed"):
-        super().__init__(message)
+class RasterScanException(DataCollectionException):
+    def __init__(self, message: str = "Data collection failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
 
     def __str__(self) -> str:
         return self.message
 
 
-class AutoRasterSampleSkipped(Exception):
+class AutoRasterSampleSkipped(AutomationError):
     """Raised when automation should skip the current sample because auto-raster is too large."""
 
 
-class LoopCenteringFailed(Exception):
-    def __init__(self, message: str = "Loop Centering did not detect a sample"):
-        super().__init__(message)
+class LoopCenteringFailed(AutomationError):
+    def __init__(self, message: str = "Loop Centering did not detect a sample", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -69,9 +137,11 @@ class LoopCenteringFailed(Exception):
         return self.message
 
 
-class UnmountingFailed(Exception):
-    def __init__(self, message: str = "A sample was not unmounted"):
-        super().__init__(message)
+class UnmountingFailed(TellException):
+    """Sample failed to unmount. Optionally critical via instance flag."""
+
+    def __init__(self, message: str = "A sample was not unmounted", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -79,9 +149,14 @@ class UnmountingFailed(Exception):
         return self.message
 
 
-class MountingFailed(Exception):
-    def __init__(self, message: str = "A sample was not mounted"):
-        super().__init__(message)
+class MountingFailed(TellException):
+    """Sample failed to mount. Optionally critical via instance flag.
+
+    Absorbs ``TellMountFailedException`` (still present for back-compat during
+    migration; will be removed in Phase 3)."""
+
+    def __init__(self, message: str = "A sample was not mounted", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -89,10 +164,10 @@ class MountingFailed(Exception):
         return self.message
 
 
-class ManualMountException(Exception):
+class ManualMountException(AareUserError):
     """Custom exception for manual mounting"""
-    def __init__(self, message: str = "Manual mounting failed"):
-        super().__init__(message)
+    def __init__(self, message: str = "Manual mounting failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -100,10 +175,10 @@ class ManualMountException(Exception):
         return self.message
 
 
-class SmartMagnetFaultException(Exception):
+class SmartMagnetFaultException(AutomationError):
     """Custom exception for smart magnet fault"""
-    def __init__(self, message: str = "Smart magnet fault"):
-        super().__init__(message)
+    def __init__(self, message: str = "Smart magnet fault", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -111,11 +186,13 @@ class SmartMagnetFaultException(Exception):
         return self.message
 
 
-class TellMountFailedException(Exception):
-    """Custom exception for mount failure"""
+class TellMountFailedException(TellException):
+    """Deprecated -- will be absorbed by MountingFailed in Phase 3.
 
-    def __init__(self, message: str = "Tell mount failed"):
-        super().__init__(message)
+    Currently kept as a TellException family member for back-compat."""
+
+    def __init__(self, message: str = "Tell mount failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -123,11 +200,11 @@ class TellMountFailedException(Exception):
         return self.message
 
 
-class TellCommandWhileBusyException(Exception):
+class TellCommandWhileBusyException(TellException):
     """Custom exception for trying to move Tell when it is busy"""
 
-    def __init__(self, message: str = "Tell is busy"):
-        super().__init__(message)
+    def __init__(self, message: str = "Tell is busy", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -135,11 +212,11 @@ class TellCommandWhileBusyException(Exception):
         return self.message
 
 
-class TellConnectionException(Exception):
+class TellConnectionException(TellException):
     """Custom exception for connection problems"""
 
-    def __init__(self, message: str = "Lost connection to Tell"):
-        super().__init__(message)
+    def __init__(self, message: str = "Lost connection to Tell", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -147,9 +224,9 @@ class TellConnectionException(Exception):
         return self.message
 
 
-class WarningTellException(Exception):
-    def __init__(self, message: str = "Warning error in TELL"):
-        super().__init__(message)
+class WarningTellException(TellException):
+    def __init__(self, message: str = "Warning error in TELL", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -157,9 +234,9 @@ class WarningTellException(Exception):
         return self.message
 
 
-class CriticalTellException(Exception):
-    def __init__(self, message: str = "Critical error in TELL"):
-        super().__init__(message)
+class CriticalTellException(TellException):
+    def __init__(self, message: str = "Critical error in TELL", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -167,9 +244,9 @@ class CriticalTellException(Exception):
         return self.message
 
 
-class AXCFailed(Exception):
-    def __init__(self, message: str = "Auto X-ray centering failed"):
-        super().__init__(message)
+class AXCFailed(AutomationError):
+    def __init__(self, message: str = "Auto X-ray centering failed", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -177,9 +254,9 @@ class AXCFailed(Exception):
         return self.message
 
 
-class BeamlineBusyException(Exception):
-    def __init__(self, message: str = "Beamline is in busy state"):
-        super().__init__(message)
+class BeamlineBusyException(BeamlineStateException):
+    def __init__(self, message: str = "Beamline is in busy state", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -187,9 +264,9 @@ class BeamlineBusyException(Exception):
         return self.message
 
 
-class BeamlineBusyTimeoutException(Exception):
-    def __init__(self, message: str = "Beamline busy state expired during operation"):
-        super().__init__(message)
+class BeamlineBusyTimeoutException(BeamlineStateException):
+    def __init__(self, message: str = "Beamline busy state expired during operation", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -197,9 +274,15 @@ class BeamlineBusyTimeoutException(Exception):
         return self.message
 
 
-class SampleException(Exception):
-    def __init__(self, message: str = "Sample not found"):
-        super().__init__(message)
+class SampleException(AareUserError):
+    """Requested sample could not be located.
+
+    NOTE: not listed in the redesign hierarchy (§1.2); parented under
+    AareUserError pending confirmation -- "sample not found" reads as bad
+    input rather than a runtime failure."""
+
+    def __init__(self, message: str = "Sample not found", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(message, extra={"exception:": Exception})
 
@@ -207,14 +290,19 @@ class SampleException(Exception):
         return self.message
 
 
-class AuthenticationException(Exception):
+# ---------------------------------------------------------------------------
+# Auth exceptions
+# ---------------------------------------------------------------------------
+
+class AuthenticationException(AareAuthError):
     def __init__(self,
                  message: str = "Authentication failed.",
                  *,
                  status_code: int = 401,
                  headers: dict[str, str] | None = None,
-                 code: AuthErrorCode = AuthErrorCode.AUTHENTICATION_FAILED):
-        super().__init__(message)
+                 code: AuthErrorCode = AuthErrorCode.AUTHENTICATION_FAILED,
+                 critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         self.status_code = status_code
         self.headers = headers
@@ -225,16 +313,18 @@ class AuthenticationException(Exception):
         return self.message
 
 
-class UserRightsException(Exception):
+class UserRightsException(AareAuthError):
     _last_log_ts_by_message: dict[str, float] = {}
     _throttle_window_s = 30.0
+
     def __init__(self,
                  message: str = "User does not have rights to perform this action.",
                  *,
                  status_code: int = 403,
                  headers: dict[str, str] | None = None,
-                 code: AuthErrorCode = AuthErrorCode.FORBIDDEN):
-        super().__init__(message)
+                 code: AuthErrorCode = AuthErrorCode.FORBIDDEN,
+                 critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         self.status_code = status_code
         self.headers = headers
@@ -250,7 +340,11 @@ class UserRightsException(Exception):
         return self.message
 
 
-class SmargonCommunicationError(Exception):
+# ---------------------------------------------------------------------------
+# Device communication exceptions
+# ---------------------------------------------------------------------------
+
+class SmargonCommunicationError(SmargonException):
     """
     Raised when Smargon HTTP communication fails (connection refused, timeout, bad HTTP status, etc).
     Keep the original exception in `__cause__` by using `raise ... from e`.
@@ -262,10 +356,11 @@ class SmargonCommunicationError(Exception):
         *,
         endpoint: str | None = None,
         base_url: str | None = None,
-        operation: str | None = None,  # e.g. "GET" / "PUT"
+        operation: str | None = None,
         status_code: int | None = None,
+        critical: bool | None = None,
     ):
-        super().__init__(message)
+        super().__init__(message, critical=critical)
         self.message = message
         self.endpoint = endpoint
         self.base_url = base_url
@@ -286,7 +381,7 @@ class SmargonCommunicationError(Exception):
         return self.message
 
 
-class TellCommunicationError(Exception):
+class TellCommunicationError(TellException):
     """
     Raised when TELL HTTP/PShell communication fails (timeouts, connection refused, etc).
     Intended to be caught centrally by FastAPI exception handlers.
@@ -298,9 +393,10 @@ class TellCommunicationError(Exception):
         *,
         endpoint: str | None = None,
         base_url: str | None = None,
-        operation: str | None = None,  # e.g. "GET"
+        operation: str | None = None,
+        critical: bool | None = None,
     ):
-        super().__init__(message)
+        super().__init__(message, critical=critical)
         self.message = message
         self.endpoint = endpoint
         self.base_url = base_url
@@ -318,7 +414,8 @@ class TellCommunicationError(Exception):
     def __str__(self) -> str:
         return self.message
 
-class JFJochCommunicationError(Exception):
+
+class JFJochCommunicationError(JFJochException):
     """
     Raised when JFJoch HTTP/API communication fails.
     Intended for scan-time fallbacks and GUI-visible alerts.
@@ -332,8 +429,9 @@ class JFJochCommunicationError(Exception):
         endpoint: str | None = None,
         base_url: str | None = None,
         status_code: int | None = None,
+        critical: bool | None = None,
     ):
-        super().__init__(message)
+        super().__init__(message, critical=critical)
         self.message = message
         self.operation = operation
         self.endpoint = endpoint
@@ -353,8 +451,13 @@ class JFJochCommunicationError(Exception):
     def __str__(self) -> str:
         return self.message
 
-class AareDBCommunicationError(Exception):
-    """Raised when AareDB HTTPS communication fails."""
+
+class AareDBCommunicationError(AutomationError):
+    """Raised when AareDB HTTPS communication fails.
+
+    Default not critical -- DB hiccups don't always halt automation. Raise with
+    ``critical=True`` at call sites where a DB read failure should escalate."""
+
     def __init__(
         self,
         message: str = "AareDB communication error",
@@ -363,8 +466,9 @@ class AareDBCommunicationError(Exception):
         endpoint: str | None = None,
         base_url: str | None = None,
         status_code: int | None = None,
+        critical: bool | None = None,
     ):
-        super().__init__(message)
+        super().__init__(message, critical=critical)
         self.message = message
         self.operation = operation
         self.endpoint = endpoint
@@ -384,7 +488,8 @@ class AareDBCommunicationError(Exception):
     def __str__(self) -> str:
         return self.message
 
-class AerotechCommunicationError(Exception):
+
+class AerotechCommunicationError(AerotechException):
     """
     Raised when Aerotech HTTP/API communication fails (connection refused, timeout, bad HTTP status, etc).
     Keep the original exception in `__cause__` by using `raise ... from e`.
@@ -398,8 +503,9 @@ class AerotechCommunicationError(Exception):
         base_url: str | None = None,
         operation: str | None = None,
         status_code: int | None = None,
+        critical: bool | None = None,
     ):
-        super().__init__(message)
+        super().__init__(message, critical=critical)
         self.message = message
         self.endpoint = endpoint
         self.base_url = base_url
@@ -419,9 +525,10 @@ class AerotechCommunicationError(Exception):
     def __str__(self) -> str:
         return self.message
 
-class MagnetPositionSensorErorr(Exception):
-    def __init__(self, message: str = "Magnet position sensor error"):
-        super().__init__(message)
+
+class MagnetPositionSensorErorr(AutomationError):
+    def __init__(self, message: str = "Magnet position sensor error", *, critical: bool | None = None):
+        super().__init__(message, critical=critical)
         self.message = message
         logger.error(
             message,
@@ -431,7 +538,8 @@ class MagnetPositionSensorErorr(Exception):
     def __str__(self) -> str:
         return self.message
 
-class BECCommunicationError(Exception):
+
+class BECCommunicationError(BECException):
     def __init__(
             self,
             message: str = "BEC communication error",
@@ -440,7 +548,9 @@ class BECCommunicationError(Exception):
             operation: str | None = None,
             endpoint: str | None = None,
             base_url: str | None = None,
+            critical: bool | None = None,
             ):
+        super().__init__(message, critical=critical)
         self.message = message
         self.operation = operation
         self.exception = exception