letta-client 0.1.318__py3-none-any.whl → 0.1.320__py3-none-any.whl

letta_client/agents/messages/raw_client.py

@@ -30,7 +30,7 @@ from .types.letta_streaming_response import LettaStreamingResponse
 from .types.message_search_request_search_mode import MessageSearchRequestSearchMode
 from .types.messages_modify_request import MessagesModifyRequest
 from .types.messages_modify_response import MessagesModifyResponse
-from .types.messages_preview_raw_payload_request import MessagesPreviewRawPayloadRequest
+from .types.messages_preview_request import MessagesPreviewRequest
 
 # this is used as the default value for optional parameters
 OMIT = typing.cast(typing.Any, ...)
@@ -498,13 +498,14 @@ class RawMessagesClient:
         search_mode: typing.Optional[MessageSearchRequestSearchMode] = OMIT,
         roles: typing.Optional[typing.Sequence[MessageRole]] = OMIT,
         project_id: typing.Optional[str] = OMIT,
+        template_id: typing.Optional[str] = OMIT,
         limit: typing.Optional[int] = OMIT,
         start_date: typing.Optional[dt.datetime] = OMIT,
         end_date: typing.Optional[dt.datetime] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> HttpResponse[typing.List[MessageSearchResult]]:
         """
-        Search messages across the entire organization with optional project filtering. Returns messages with FTS/vector ranks and total RRF score.
+        Search messages across the entire organization with optional project and template filtering. Returns messages with FTS/vector ranks and total RRF score.
 
         This is a cloud-only feature.
 
@@ -522,6 +523,9 @@ class RawMessagesClient:
         project_id : typing.Optional[str]
             Filter messages by project ID
 
+        template_id : typing.Optional[str]
+            Filter messages by template ID
+
         limit : typing.Optional[int]
             Maximum number of results to return
 
@@ -547,6 +551,7 @@ class RawMessagesClient:
                 "search_mode": search_mode,
                 "roles": roles,
                 "project_id": project_id,
+                "template_id": template_id,
                 "limit": limit,
                 "start_date": start_date,
                 "end_date": end_date,
@@ -746,12 +751,8 @@ class RawMessagesClient:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
 
-    def preview_raw_payload(
-        self,
-        agent_id: str,
-        *,
-        request: MessagesPreviewRawPayloadRequest,
-        request_options: typing.Optional[RequestOptions] = None,
+    def preview(
+        self, agent_id: str, *, request: MessagesPreviewRequest, request_options: typing.Optional[RequestOptions] = None
     ) -> HttpResponse[typing.Dict[str, typing.Optional[typing.Any]]]:
         """
         Inspect the raw LLM request payload without sending it.
@@ -764,7 +765,7 @@ class RawMessagesClient:
         ----------
         agent_id : str
 
-        request : MessagesPreviewRawPayloadRequest
+        request : MessagesPreviewRequest
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -778,7 +779,7 @@ class RawMessagesClient:
             f"v1/agents/{jsonable_encoder(agent_id)}/messages/preview-raw-payload",
             method="POST",
             json=convert_and_respect_annotation_metadata(
-                object_=request, annotation=MessagesPreviewRawPayloadRequest, direction="write"
+                object_=request, annotation=MessagesPreviewRequest, direction="write"
             ),
             headers={
                 "content-type": "application/json",
@@ -812,6 +813,56 @@ class RawMessagesClient:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
 
+    def summarize(
+        self, agent_id: str, *, max_message_length: int, request_options: typing.Optional[RequestOptions] = None
+    ) -> HttpResponse[None]:
+        """
+        Summarize an agent's conversation history to a target message length.
+
+        This endpoint summarizes the current message history for a given agent,
+        truncating and compressing it down to the specified `max_message_length`.
+
+        Parameters
+        ----------
+        agent_id : str
+
+        max_message_length : int
+            Maximum number of messages to retain after summarization.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        HttpResponse[None]
+        """
+        _response = self._client_wrapper.httpx_client.request(
+            f"v1/agents/{jsonable_encoder(agent_id)}/summarize",
+            method="POST",
+            params={
+                "max_message_length": max_message_length,
+            },
+            request_options=request_options,
+        )
+        try:
+            if 200 <= _response.status_code < 300:
+                return HttpResponse(response=_response, data=None)
+            if _response.status_code == 422:
+                raise UnprocessableEntityError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        HttpValidationError,
+                        construct_type(
+                            type_=HttpValidationError,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
+        raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
 
 class AsyncRawMessagesClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
@@ -1275,13 +1326,14 @@ class AsyncRawMessagesClient:
         search_mode: typing.Optional[MessageSearchRequestSearchMode] = OMIT,
         roles: typing.Optional[typing.Sequence[MessageRole]] = OMIT,
         project_id: typing.Optional[str] = OMIT,
+        template_id: typing.Optional[str] = OMIT,
         limit: typing.Optional[int] = OMIT,
         start_date: typing.Optional[dt.datetime] = OMIT,
         end_date: typing.Optional[dt.datetime] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> AsyncHttpResponse[typing.List[MessageSearchResult]]:
         """
-        Search messages across the entire organization with optional project filtering. Returns messages with FTS/vector ranks and total RRF score.
+        Search messages across the entire organization with optional project and template filtering. Returns messages with FTS/vector ranks and total RRF score.
 
         This is a cloud-only feature.
 
@@ -1299,6 +1351,9 @@ class AsyncRawMessagesClient:
         project_id : typing.Optional[str]
             Filter messages by project ID
 
+        template_id : typing.Optional[str]
+            Filter messages by template ID
+
         limit : typing.Optional[int]
             Maximum number of results to return
 
@@ -1324,6 +1379,7 @@ class AsyncRawMessagesClient:
                 "search_mode": search_mode,
                 "roles": roles,
                 "project_id": project_id,
+                "template_id": template_id,
                 "limit": limit,
                 "start_date": start_date,
                 "end_date": end_date,
@@ -1523,12 +1579,8 @@ class AsyncRawMessagesClient:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
 
-    async def preview_raw_payload(
-        self,
-        agent_id: str,
-        *,
-        request: MessagesPreviewRawPayloadRequest,
-        request_options: typing.Optional[RequestOptions] = None,
+    async def preview(
+        self, agent_id: str, *, request: MessagesPreviewRequest, request_options: typing.Optional[RequestOptions] = None
     ) -> AsyncHttpResponse[typing.Dict[str, typing.Optional[typing.Any]]]:
         """
         Inspect the raw LLM request payload without sending it.
@@ -1541,7 +1593,7 @@ class AsyncRawMessagesClient:
         ----------
         agent_id : str
 
-        request : MessagesPreviewRawPayloadRequest
+        request : MessagesPreviewRequest
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -1555,7 +1607,7 @@ class AsyncRawMessagesClient:
             f"v1/agents/{jsonable_encoder(agent_id)}/messages/preview-raw-payload",
             method="POST",
             json=convert_and_respect_annotation_metadata(
-                object_=request, annotation=MessagesPreviewRawPayloadRequest, direction="write"
+                object_=request, annotation=MessagesPreviewRequest, direction="write"
             ),
             headers={
                 "content-type": "application/json",
@@ -1588,3 +1640,53 @@ class AsyncRawMessagesClient:
         except JSONDecodeError:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
+    async def summarize(
+        self, agent_id: str, *, max_message_length: int, request_options: typing.Optional[RequestOptions] = None
+    ) -> AsyncHttpResponse[None]:
+        """
+        Summarize an agent's conversation history to a target message length.
+
+        This endpoint summarizes the current message history for a given agent,
+        truncating and compressing it down to the specified `max_message_length`.
+
+        Parameters
+        ----------
+        agent_id : str
+
+        max_message_length : int
+            Maximum number of messages to retain after summarization.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        AsyncHttpResponse[None]
+        """
+        _response = await self._client_wrapper.httpx_client.request(
+            f"v1/agents/{jsonable_encoder(agent_id)}/summarize",
+            method="POST",
+            params={
+                "max_message_length": max_message_length,
+            },
+            request_options=request_options,
+        )
+        try:
+            if 200 <= _response.status_code < 300:
+                return AsyncHttpResponse(response=_response, data=None)
+            if _response.status_code == 422:
+                raise UnprocessableEntityError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        HttpValidationError,
+                        construct_type(
+                            type_=HttpValidationError,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
+        raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)

letta_client/agents/messages/types/__init__.py

@@ -7,7 +7,7 @@ from .letta_streaming_response import LettaStreamingResponse
 from .message_search_request_search_mode import MessageSearchRequestSearchMode
 from .messages_modify_request import MessagesModifyRequest
 from .messages_modify_response import MessagesModifyResponse
-from .messages_preview_raw_payload_request import MessagesPreviewRawPayloadRequest
+from .messages_preview_request import MessagesPreviewRequest
 
 __all__ = [
     "LettaAsyncRequestMessagesItem",
@@ -15,5 +15,5 @@ __all__ = [
     "MessageSearchRequestSearchMode",
     "MessagesModifyRequest",
     "MessagesModifyResponse",
-    "MessagesPreviewRawPayloadRequest",
+    "MessagesPreviewRequest",
 ]

letta_client/agents/messages/types/{messages_preview_raw_payload_request.py → messages_preview_request.py}

@@ -5,4 +5,4 @@ import typing
 from ....types.letta_request import LettaRequest
 from ....types.letta_streaming_request import LettaStreamingRequest
 
-MessagesPreviewRawPayloadRequest = typing.Union[LettaRequest, LettaStreamingRequest]
+MessagesPreviewRequest = typing.Union[LettaRequest, LettaStreamingRequest]