From a5558f8fcc3cdfbfd12aad1aafa14ac4e362a892 Mon Sep 17 00:00:00 2001
From: Benjamin <benjaminx@gmail.com>
Date: Thu, 7 Nov 2024 18:07:05 +0800
Subject: [PATCH 01/14] =?UTF-8?q?fix(conversation-service):=20return=20suc?=
 =?UTF-8?q?cess=20response=20after=20conversation=E2=80=A6=20(#10416)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 api/controllers/service_api/app/conversation.py | 3 +--
 api/services/conversation_service.py            | 3 +++
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/api/controllers/service_api/app/conversation.py b/api/controllers/service_api/app/conversation.py
index 527ef4ecd3..a0bd1b7412 100644
--- a/api/controllers/service_api/app/conversation.py
+++ b/api/controllers/service_api/app/conversation.py
@@ -58,10 +58,9 @@ class ConversationDetailApi(Resource):
         conversation_id = str(c_id)
 
         try:
-            ConversationService.delete(app_model, conversation_id, end_user)
+            return ConversationService.delete(app_model, conversation_id, end_user)
         except services.errors.conversation.ConversationNotExistsError:
             raise NotFound("Conversation Not Exists.")
-        return {"result": "success"}, 200
 
 
 class ConversationRenameApi(Resource):
diff --git a/api/services/conversation_service.py b/api/services/conversation_service.py
index 7bfe59afa0..ac9f577c32 100644
--- a/api/services/conversation_service.py
+++ b/api/services/conversation_service.py
@@ -160,4 +160,7 @@ class ConversationService:
         conversation = cls.get_conversation(app_model, conversation_id, user)
 
         conversation.is_deleted = True
+        conversation.updated_at = datetime.now(timezone.utc).replace(tzinfo=None)
         db.session.commit()
+
+        return {"result": "success"}, 200

From 59f8d116af6e4b8856ff95f776073d948d4f7e22 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E9=9D=9E=E6=B3=95=E6=93=8D=E4=BD=9C?= <hjlarry@163.com>
Date: Thu, 7 Nov 2024 18:10:41 +0800
Subject: [PATCH 02/14] chore: improve custom tool's display (#10410)

---
 .../tools/edit-custom-collection-modal/config-credentials.tsx | 4 ++--
 .../tools/edit-custom-collection-modal/get-schema.tsx         | 4 ++--
 .../components/tools/edit-custom-collection-modal/index.tsx   | 4 ++--
 web/app/components/tools/provider/custom-create-card.tsx      | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/web/app/components/tools/edit-custom-collection-modal/config-credentials.tsx b/web/app/components/tools/edit-custom-collection-modal/config-credentials.tsx
index d580c00102..4b24bcb931 100644
--- a/web/app/components/tools/edit-custom-collection-modal/config-credentials.tsx
+++ b/web/app/components/tools/edit-custom-collection-modal/config-credentials.tsx
@@ -120,7 +120,7 @@ const ConfigCredential: FC<Props> = ({
                   <input
                     value={tempCredential.api_key_header}
                     onChange={e => setTempCredential({ ...tempCredential, api_key_header: e.target.value })}
-                    className='w-full h-10 px-3 text-sm font-normal bg-gray-100 rounded-lg grow'
+                    className='w-full h-10 px-3 text-sm font-normal  border border-transparent bg-gray-100 rounded-lg grow outline-none focus:bg-components-input-bg-active focus:border-components-input-border-active focus:shadow-xs'
                     placeholder={t('tools.createTool.authMethod.types.apiKeyPlaceholder')!}
                   />
                 </div>
@@ -129,7 +129,7 @@ const ConfigCredential: FC<Props> = ({
                   <input
                     value={tempCredential.api_key_value}
                     onChange={e => setTempCredential({ ...tempCredential, api_key_value: e.target.value })}
-                    className='w-full h-10 px-3 text-sm font-normal bg-gray-100 rounded-lg grow'
+                    className='w-full h-10 px-3 text-sm font-normal  border border-transparent bg-gray-100 rounded-lg grow outline-none focus:bg-components-input-bg-active focus:border-components-input-border-active focus:shadow-xs'
                     placeholder={t('tools.createTool.authMethod.types.apiValuePlaceholder')!}
                   />
                 </div>
diff --git a/web/app/components/tools/edit-custom-collection-modal/get-schema.tsx b/web/app/components/tools/edit-custom-collection-modal/get-schema.tsx
index 7b0244e3e3..2552c67568 100644
--- a/web/app/components/tools/edit-custom-collection-modal/get-schema.tsx
+++ b/web/app/components/tools/edit-custom-collection-modal/get-schema.tsx
@@ -70,7 +70,7 @@ const GetSchema: FC<Props> = ({
             <div className='relative'>
               <input
                 type='text'
-                className='w-[244px] h-8 pl-1.5 pr-[44px] overflow-x-auto border border-gray-200 rounded-lg text-[13px]'
+                className='w-[244px] h-8 pl-1.5 pr-[44px] overflow-x-auto border border-gray-200 rounded-lg text-[13px] focus:outline-none focus:border-components-input-border-active'
                 placeholder={t('tools.createTool.importFromUrlPlaceHolder')!}
                 value={importUrl}
                 onChange={e => setImportUrl(e.target.value)}
@@ -89,7 +89,7 @@ const GetSchema: FC<Props> = ({
           </div>
         )}
       </div>
-      <div className='relative' ref={showExamplesRef}>
+      <div className='relative -mt-0.5' ref={showExamplesRef}>
         <Button
           size='small'
           className='space-x-1'
diff --git a/web/app/components/tools/edit-custom-collection-modal/index.tsx b/web/app/components/tools/edit-custom-collection-modal/index.tsx
index c734c91e13..555fd0dd08 100644
--- a/web/app/components/tools/edit-custom-collection-modal/index.tsx
+++ b/web/app/components/tools/edit-custom-collection-modal/index.tsx
@@ -186,8 +186,8 @@ const EditCustomCollectionModal: FC<Props> = ({
         positionCenter={isAdd && !positionLeft}
         onHide={onHide}
         title={t(`tools.createTool.${isAdd ? 'title' : 'editTitle'}`)!}
-        panelClassName='mt-2 !w-[630px]'
-        maxWidthClassName='!max-w-[630px]'
+        panelClassName='mt-2 !w-[640px]'
+        maxWidthClassName='!max-w-[640px]'
         height='calc(100vh - 16px)'
         headerClassName='!border-b-black/5'
         body={
diff --git a/web/app/components/tools/provider/custom-create-card.tsx b/web/app/components/tools/provider/custom-create-card.tsx
index 99a2ad2be3..d6aa9ab533 100644
--- a/web/app/components/tools/provider/custom-create-card.tsx
+++ b/web/app/components/tools/provider/custom-create-card.tsx
@@ -27,8 +27,8 @@ const Contribute = ({ onRefreshData }: Props) => {
 
   const linkUrl = useMemo(() => {
     if (language.startsWith('zh_'))
-      return 'https://docs.dify.ai/v/zh-hans/guides/gong-ju/quick-tool-integration'
-    return 'https://docs.dify.ai/tutorials/quick-tool-integration'
+      return 'https://docs.dify.ai/zh-hans/guides/tools#ru-he-chuang-jian-zi-ding-yi-gong-ju'
+    return 'https://docs.dify.ai/guides/tools#how-to-create-custom-tools'
   }, [language])
 
   const [isShowEditCollectionToolModal, setIsShowEditCustomCollectionModal] = useState(false)

From f562a882498d0928673354bf89ec3a08f884dad4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E8=B5=B0=E5=9C=A8=E4=BF=AE=E8=A1=8C=E7=9A=84=E5=A4=A7?=
 =?UTF-8?q?=E8=A1=97=E4=B8=8A?= <hgnulb@163.com>
Date: Thu, 7 Nov 2024 18:11:25 +0800
Subject: [PATCH 03/14] feat(Tools): add lark tools (#10117)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Co-authored-by: 黎斌 <libin.23@bytedance.com>
---
 api/core/tools/provider/_position.yaml        |   7 +
 .../feishu_base/tools/list_tables.yaml        |   2 +-
 .../feishu_base/tools/search_records.yaml     |   2 +-
 .../feishu_calendar/tools/list_events.yaml    |   2 +-
 .../feishu_calendar/tools/search_events.yaml  |   2 +-
 .../tools/list_document_blocks.yaml           |   2 +-
 .../tools/get_chat_messages.yaml              |   2 +-
 .../tools/get_thread_messages.yaml            |   2 +-
 .../feishu_spreadsheet/tools/add_cols.yaml    |   2 +-
 .../feishu_spreadsheet/tools/add_rows.yaml    |   2 +-
 .../feishu_spreadsheet/tools/read_cols.yaml   |   4 +-
 .../feishu_spreadsheet/tools/read_rows.yaml   |   4 +-
 .../feishu_spreadsheet/tools/read_table.yaml  |   4 +-
 .../feishu_wiki/tools/get_wiki_nodes.yaml     |   2 +-
 .../builtin/lark_base/_assets/icon.png        | Bin 0 -> 2918 bytes
 .../provider/builtin/lark_base/lark_base.py   |   7 +
 .../provider/builtin/lark_base/lark_base.yaml |  36 +
 .../builtin/lark_base/tools/add_records.py    |  21 +
 .../builtin/lark_base/tools/add_records.yaml  |  91 ++
 .../builtin/lark_base/tools/create_base.py    |  18 +
 .../builtin/lark_base/tools/create_base.yaml  |  42 +
 .../builtin/lark_base/tools/create_table.py   |  20 +
 .../builtin/lark_base/tools/create_table.yaml |  61 ++
 .../builtin/lark_base/tools/delete_records.py |  20 +
 .../lark_base/tools/delete_records.yaml       |  86 ++
 .../builtin/lark_base/tools/delete_tables.py  |  19 +
 .../lark_base/tools/delete_tables.yaml        |  49 ++
 .../builtin/lark_base/tools/get_base_info.py  |  17 +
 .../lark_base/tools/get_base_info.yaml        |  23 +
 .../builtin/lark_base/tools/list_tables.py    |  19 +
 .../builtin/lark_base/tools/list_tables.yaml  |  50 ++
 .../builtin/lark_base/tools/read_records.py   |  21 +
 .../builtin/lark_base/tools/read_records.yaml |  86 ++
 .../builtin/lark_base/tools/search_records.py |  39 +
 .../lark_base/tools/search_records.yaml       | 163 ++++
 .../builtin/lark_base/tools/update_records.py |  21 +
 .../lark_base/tools/update_records.yaml       |  91 ++
 .../builtin/lark_calendar/_assets/icon.png    | Bin 0 -> 5488 bytes
 .../builtin/lark_calendar/lark_calendar.py    |   7 +
 .../builtin/lark_calendar/lark_calendar.yaml  |  36 +
 .../tools/add_event_attendees.py              |  20 +
 .../tools/add_event_attendees.yaml            |  54 ++
 .../lark_calendar/tools/create_event.py       |  26 +
 .../lark_calendar/tools/create_event.yaml     | 119 +++
 .../lark_calendar/tools/delete_event.py       |  19 +
 .../lark_calendar/tools/delete_event.yaml     |  38 +
 .../tools/get_primary_calendar.py             |  18 +
 .../tools/get_primary_calendar.yaml           |  37 +
 .../lark_calendar/tools/list_events.py        |  21 +
 .../lark_calendar/tools/list_events.yaml      |  62 ++
 .../lark_calendar/tools/search_events.py      |  23 +
 .../lark_calendar/tools/search_events.yaml    | 100 +++
 .../lark_calendar/tools/update_event.py       |  24 +
 .../lark_calendar/tools/update_event.yaml     | 100 +++
 .../builtin/lark_document/_assets/icon.svg    |   9 +
 .../builtin/lark_document/lark_document.py    |   7 +
 .../builtin/lark_document/lark_document.yaml  |  36 +
 .../lark_document/tools/create_document.py    |  19 +
 .../lark_document/tools/create_document.yaml  |  48 +
 .../tools/get_document_content.py             |  19 +
 .../tools/get_document_content.yaml           |  70 ++
 .../tools/list_document_blocks.py             |  20 +
 .../tools/list_document_blocks.yaml           |  74 ++
 .../lark_document/tools/write_document.py     |  19 +
 .../lark_document/tools/write_document.yaml   |  57 ++
 .../lark_message_and_group/_assets/icon.png   | Bin 0 -> 4482 bytes
 .../lark_message_and_group.py                 |   7 +
 .../lark_message_and_group.yaml               |  36 +
 .../tools/get_chat_messages.py                |  23 +
 .../tools/get_chat_messages.yaml              |  96 ++
 .../tools/get_thread_messages.py              |  21 +
 .../tools/get_thread_messages.yaml            |  72 ++
 .../tools/send_bot_message.py                 |  20 +
 .../tools/send_bot_message.yaml               | 125 +++
 .../tools/send_webhook_message.py             |  19 +
 .../tools/send_webhook_message.yaml           |  68 ++
 .../builtin/lark_spreadsheet/_assets/icon.png | Bin 0 -> 5417 bytes
 .../lark_spreadsheet/lark_spreadsheet.py      |   7 +
 .../lark_spreadsheet/lark_spreadsheet.yaml    |  36 +
 .../lark_spreadsheet/tools/add_cols.py        |  22 +
 .../lark_spreadsheet/tools/add_cols.yaml      |  72 ++
 .../lark_spreadsheet/tools/add_rows.py        |  22 +
 .../lark_spreadsheet/tools/add_rows.yaml      |  72 ++
 .../tools/create_spreadsheet.py               |  19 +
 .../tools/create_spreadsheet.yaml             |  35 +
 .../lark_spreadsheet/tools/get_spreadsheet.py |  19 +
 .../tools/get_spreadsheet.yaml                |  49 ++
 .../tools/list_spreadsheet_sheets.py          |  18 +
 .../tools/list_spreadsheet_sheets.yaml        |  23 +
 .../lark_spreadsheet/tools/read_cols.py       |  23 +
 .../lark_spreadsheet/tools/read_cols.yaml     |  97 +++
 .../lark_spreadsheet/tools/read_rows.py       |  23 +
 .../lark_spreadsheet/tools/read_rows.yaml     |  97 +++
 .../lark_spreadsheet/tools/read_table.py      |  23 +
 .../lark_spreadsheet/tools/read_table.yaml    | 122 +++
 .../builtin/lark_task/_assets/icon.png        | Bin 0 -> 2115 bytes
 .../provider/builtin/lark_task/lark_task.py   |   7 +
 .../provider/builtin/lark_task/lark_task.yaml |  36 +
 .../builtin/lark_task/tools/add_members.py    |  20 +
 .../builtin/lark_task/tools/add_members.yaml  |  58 ++
 .../builtin/lark_task/tools/create_task.py    |  22 +
 .../builtin/lark_task/tools/create_task.yaml  |  74 ++
 .../builtin/lark_task/tools/delete_task.py    |  18 +
 .../builtin/lark_task/tools/delete_task.yaml  |  24 +
 .../builtin/lark_task/tools/update_task.py    |  23 +
 .../builtin/lark_task/tools/update_task.yaml  |  89 ++
 .../builtin/lark_wiki/_assets/icon.png        | Bin 0 -> 2827 bytes
 .../provider/builtin/lark_wiki/lark_wiki.py   |   7 +
 .../provider/builtin/lark_wiki/lark_wiki.yaml |  36 +
 .../builtin/lark_wiki/tools/get_wiki_nodes.py |  21 +
 .../lark_wiki/tools/get_wiki_nodes.yaml       |  63 ++
 api/core/tools/utils/feishu_api_utils.py      | 132 ++-
 api/core/tools/utils/lark_api_utils.py        | 820 ++++++++++++++++++
 113 files changed, 4788 insertions(+), 49 deletions(-)
 create mode 100644 api/core/tools/provider/builtin/lark_base/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_base/lark_base.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/lark_base.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/add_records.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/add_records.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/create_base.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/create_base.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/create_table.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/create_table.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/delete_records.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/delete_records.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/delete_tables.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/delete_tables.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/get_base_info.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/get_base_info.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/list_tables.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/list_tables.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/read_records.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/read_records.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/search_records.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/search_records.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/update_records.py
 create mode 100644 api/core/tools/provider/builtin/lark_base/tools/update_records.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/lark_calendar.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/lark_calendar.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/create_event.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/create_event.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/delete_event.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/delete_event.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/list_events.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/list_events.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/search_events.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/search_events.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/update_event.py
 create mode 100644 api/core/tools/provider/builtin/lark_calendar/tools/update_event.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_document/_assets/icon.svg
 create mode 100644 api/core/tools/provider/builtin/lark_document/lark_document.py
 create mode 100644 api/core/tools/provider/builtin/lark_document/lark_document.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/create_document.py
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/create_document.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/get_document_content.py
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/get_document_content.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.py
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/write_document.py
 create mode 100644 api/core/tools/provider/builtin/lark_document/tools/write_document.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.py
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.py
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.py
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.py
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.py
 create mode 100644 api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.py
 create mode 100644 api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_task/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_task/lark_task.py
 create mode 100644 api/core/tools/provider/builtin/lark_task/lark_task.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/add_members.py
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/add_members.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/create_task.py
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/create_task.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/delete_task.py
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/delete_task.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/update_task.py
 create mode 100644 api/core/tools/provider/builtin/lark_task/tools/update_task.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_wiki/_assets/icon.png
 create mode 100644 api/core/tools/provider/builtin/lark_wiki/lark_wiki.py
 create mode 100644 api/core/tools/provider/builtin/lark_wiki/lark_wiki.yaml
 create mode 100644 api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.py
 create mode 100644 api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.yaml
 create mode 100644 api/core/tools/utils/lark_api_utils.py

diff --git a/api/core/tools/provider/_position.yaml b/api/core/tools/provider/_position.yaml
index 336588e3e2..d80974486d 100644
--- a/api/core/tools/provider/_position.yaml
+++ b/api/core/tools/provider/_position.yaml
@@ -48,6 +48,13 @@
 - feishu_task
 - feishu_calendar
 - feishu_spreadsheet
+- lark_base
+- lark_document
+- lark_message_and_group
+- lark_wiki
+- lark_task
+- lark_calendar
+- lark_spreadsheet
 - slack
 - twilio
 - wecom
diff --git a/api/core/tools/provider/builtin/feishu_base/tools/list_tables.yaml b/api/core/tools/provider/builtin/feishu_base/tools/list_tables.yaml
index 5a3891bd45..7571519039 100644
--- a/api/core/tools/provider/builtin/feishu_base/tools/list_tables.yaml
+++ b/api/core/tools/provider/builtin/feishu_base/tools/list_tables.yaml
@@ -34,7 +34,7 @@ parameters:
         Page size, default value: 20, maximum value: 100.
       zh_Hans: 分页大小，默认值：20，最大值：100。
     llm_description: 分页大小，默认值：20，最大值：100。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_base/tools/search_records.yaml b/api/core/tools/provider/builtin/feishu_base/tools/search_records.yaml
index 6cac4b0524..decf76d53e 100644
--- a/api/core/tools/provider/builtin/feishu_base/tools/search_records.yaml
+++ b/api/core/tools/provider/builtin/feishu_base/tools/search_records.yaml
@@ -147,7 +147,7 @@ parameters:
         Page size, default value: 20, maximum value: 500.
       zh_Hans: 分页大小，默认值：20，最大值：500。
     llm_description: 分页大小，默认值：20，最大值：500。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_calendar/tools/list_events.yaml b/api/core/tools/provider/builtin/feishu_calendar/tools/list_events.yaml
index f4a5bfe6ba..5f0155a246 100644
--- a/api/core/tools/provider/builtin/feishu_calendar/tools/list_events.yaml
+++ b/api/core/tools/provider/builtin/feishu_calendar/tools/list_events.yaml
@@ -47,7 +47,7 @@ parameters:
       en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 50, and the value range is [50,1000].
       zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 50，取值范围为 [50,1000]。
     llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 50，取值范围为 [50,1000]。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_calendar/tools/search_events.yaml b/api/core/tools/provider/builtin/feishu_calendar/tools/search_events.yaml
index e92a282091..bd60a07b5b 100644
--- a/api/core/tools/provider/builtin/feishu_calendar/tools/search_events.yaml
+++ b/api/core/tools/provider/builtin/feishu_calendar/tools/search_events.yaml
@@ -85,7 +85,7 @@ parameters:
       en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [10,100].
       zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [10,100]。
     llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [10,100]。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_document/tools/list_document_blocks.yaml b/api/core/tools/provider/builtin/feishu_document/tools/list_document_blocks.yaml
index 5b8ef7d53c..d4fab96c1f 100644
--- a/api/core/tools/provider/builtin/feishu_document/tools/list_document_blocks.yaml
+++ b/api/core/tools/provider/builtin/feishu_document/tools/list_document_blocks.yaml
@@ -59,7 +59,7 @@ parameters:
       en_US: Paging size, the default and maximum value is 500.
       zh_Hans: 分页大小, 默认值和最大值为 500。
     llm_description: 分页大小, 表示一次请求最多返回多少条数据，默认值和最大值为 500。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_message/tools/get_chat_messages.yaml b/api/core/tools/provider/builtin/feishu_message/tools/get_chat_messages.yaml
index 153c8c80e5..984c9120e8 100644
--- a/api/core/tools/provider/builtin/feishu_message/tools/get_chat_messages.yaml
+++ b/api/core/tools/provider/builtin/feishu_message/tools/get_chat_messages.yaml
@@ -81,7 +81,7 @@ parameters:
       en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [1,50].
       zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
     llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_message/tools/get_thread_messages.yaml b/api/core/tools/provider/builtin/feishu_message/tools/get_thread_messages.yaml
index 8d5fed9d0b..85a138292f 100644
--- a/api/core/tools/provider/builtin/feishu_message/tools/get_thread_messages.yaml
+++ b/api/core/tools/provider/builtin/feishu_message/tools/get_thread_messages.yaml
@@ -57,7 +57,7 @@ parameters:
       en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [1,50].
       zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
     llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_cols.yaml b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_cols.yaml
index ef457f8e00..b73335f405 100644
--- a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_cols.yaml
+++ b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_cols.yaml
@@ -56,7 +56,7 @@ parameters:
       en_US: Number of columns to add, range (0-5000].
       zh_Hans: 要增加的列数，范围(0-5000]。
     llm_description: 要增加的列数，范围(0-5000]。
-    form: llm
+    form: form
 
   - name: values
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_rows.yaml b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_rows.yaml
index 37653325ae..6bce305b98 100644
--- a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_rows.yaml
+++ b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/add_rows.yaml
@@ -56,7 +56,7 @@ parameters:
       en_US: Number of rows to add, range (0-5000].
       zh_Hans: 要增加行数，范围(0-5000]。
     llm_description: 要增加行数，范围(0-5000]。
-    form: llm
+    form: form
 
   - name: values
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_cols.yaml b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_cols.yaml
index 3273857b70..34da74592d 100644
--- a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_cols.yaml
+++ b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_cols.yaml
@@ -82,7 +82,7 @@ parameters:
       en_US: Starting column number, starting from 1.
       zh_Hans: 起始列号，从 1 开始。
     llm_description: 起始列号，从 1 开始。
-    form: llm
+    form: form
 
   - name: num_cols
     type: number
@@ -94,4 +94,4 @@ parameters:
       en_US: Number of columns to read.
       zh_Hans: 读取列数
     llm_description: 读取列数
-    form: llm
+    form: form
diff --git a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_rows.yaml b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_rows.yaml
index 3e9206e8ef..5dfa8d5835 100644
--- a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_rows.yaml
+++ b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_rows.yaml
@@ -82,7 +82,7 @@ parameters:
       en_US: Starting row number, starting from 1.
       zh_Hans: 起始行号，从 1 开始。
     llm_description: 起始行号，从 1 开始。
-    form: llm
+    form: form
 
   - name: num_rows
     type: number
@@ -94,4 +94,4 @@ parameters:
       en_US: Number of rows to read.
       zh_Hans: 读取行数
     llm_description: 读取行数
-    form: llm
+    form: form
diff --git a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_table.yaml b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_table.yaml
index e3dc80e1eb..10534436d6 100644
--- a/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_table.yaml
+++ b/api/core/tools/provider/builtin/feishu_spreadsheet/tools/read_table.yaml
@@ -82,7 +82,7 @@ parameters:
       en_US: Starting row number, starting from 1.
       zh_Hans: 起始行号，从 1 开始。
     llm_description: 起始行号，从 1 开始。
-    form: llm
+    form: form
 
   - name: num_rows
     type: number
@@ -94,7 +94,7 @@ parameters:
       en_US: Number of rows to read.
       zh_Hans: 读取行数
     llm_description: 读取行数
-    form: llm
+    form: form
 
   - name: range
     type: string
diff --git a/api/core/tools/provider/builtin/feishu_wiki/tools/get_wiki_nodes.yaml b/api/core/tools/provider/builtin/feishu_wiki/tools/get_wiki_nodes.yaml
index 7d6ac3c824..74d51e7bcb 100644
--- a/api/core/tools/provider/builtin/feishu_wiki/tools/get_wiki_nodes.yaml
+++ b/api/core/tools/provider/builtin/feishu_wiki/tools/get_wiki_nodes.yaml
@@ -36,7 +36,7 @@ parameters:
       en_US: The size of each page, with a maximum value of 50.
       zh_Hans: 分页大小，最大值 50。
     llm_description: 分页大小，最大值 50。
-    form: llm
+    form: form
 
   - name: page_token
     type: string
diff --git a/api/core/tools/provider/builtin/lark_base/_assets/icon.png b/api/core/tools/provider/builtin/lark_base/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..036e586772ef502791c73db50681267a69bab98d
GIT binary patch
literal 2918
zcmV-s3z_tZP)<h;3K|Lk000e1NJLTq002M$002M;1^@s6s%dfF000XuNkl<ZcwX&R
zTThkQ8J?doKf!;{`3L2$>QtdB0&-LiqJV&ao0Ec4P3=W5Y}2MT?W9T5q?t~p)23rG
zHhxLdWG>pYNpE`0crlHelYoGG!)CKq-{V^ATYF2z{h8jdk|*!^c@O(b;GK7l-qAaH
zNAKtz{O0EDyVG~=d$S9+?*9#(BcAWAzNrP<Fniao0qo2-jYY}2H8Xb&$?7rEB_=+}
z(YUvX$Fh$(W|>=K%Hud@@7ZMLp6!Gm=U6;*rVpBZj7_H(teIXwDLVOZIFdA&=#0qr
zab6;n)pT7}xx)U+CHMlqin)01%-+<3NwDeE9peU=5;*<NzC)Oc(#KF2Kf&ixTzN#g
zLoWDQ&o7)@&v|#u)5*8-H9&(W*ktlH4I<zHCT|-vsq!`th_=ti*gogn!JMF`3LI}_
zyI#V{Inpoa4aoiox%qOEo8LD%Xb!r~-7b^FVb~`zTFxW3LqFEUI@!V%+NoGLM`*(V
zc&>f?QohLa2z&ruT5%(gtm7~mr(vD0IdRPmsF7Ondydeb?#ng4HslOvC@1kAZ~P{J
zY+`QDBB;?^8FT<NB;TC)=F0}5dP-8g3@SC`!y55d^q`pLGN_ltspgi{>kPp>H*S?X
z<6Yk90ckRRYtN2jGm;F|VTeHIMb$rkbI&-x#%?N)m*~S>mxr|)OYve2_lsqEF5=7P
zD?lOuF5@ufr5T$okv(>kBsO*u7o!}-tonU^DARtdMZMCmGq|3(H1djceP)Yw@ji!s
zK$g&wfa?Ta8YSt_i2XjMNAbA~;Bv8$gAwLwJ=du3`b^33Q_W0z7Ugk0XLd&7eGdJ`
zX7-ZN8Ei%a5Jnl~Ly7Uwhl6o!SM$y2bO{Hf+K?;Y^SMyx`I*DVyx;dr)dk$;58{0e
z{YGX=$;dPi!vZ6OU5j#*d<^z8*cMFfx;$3{gZW(LBYxL|4ifd@9G-)7rSFvX^!*)R
z_IRH|zv1c96IcY7Aq>M(vYfMrr*?b{{V<^SmkjGtGwt(SKIuC_jA~N3u4mAjbv#1g
zb9}^;!2u?Bl53N@=Gx?r9adlikmy&N^a{VPalWA5`;kA4hrI$l%sJ5K{IwroC<h0~
z!Jr5nIFzD84q^WOu6^{ovibD!o(*yTLCIh|#UuFWHzoV%aXD3ocABH?eDmLeW9Q&t
z!ZnmDrfkQZ%jf`uaDc%ZI|eayqi9tQiH~eK!1Fa@U-ce5EJ3cX!@+;{i&C8SKVP9>
z;BWmd_oR7>bOf}Om*yj1F@pn4>?8vdI|hNx5CaoMrz>vV+lgmQy@x;Dg}MR`N%e@&
z|5%RGUh9qDf>|b@F9Q41Y&fIL<@^~O07w%Yd|<q2u8z~-v<NnH_Bh%0x1a6Bv(o#=
zzxOH5`J9XTz3=RpgT9(-@5%4?s6OPO{shl2z6=g9UQDix6=2zdy*gGjsym-zj1l`o
z+v|9|R1NxsA7Y4zSLz`5CGiZ9E}T@ux<K{X4}c{7*9)fqIyR=}7H!hO=l!UIUt=_c
zukm#Ss^dKNIX?}pyk1u>pBMU(&(#N@c<l%1A1x%8M_-!DqkIGe+D-=k{!#X^A9aqQ
zR39wCxpLGH%FB6ZZ>RmKJgJx5n1f^%2e>@4o%D_1Qrk!>P#vY$`$j@PNT~C*7_Y9=
z(JFd`(0tCpKJiSEt+}-ypl^8l2{!D72E8NOhIN#@9Qt`6Ilu@G<84LnFxPr;9{6c^
zT}pM%_uSgM66yZEfsQ|3yw9Ou@9<UvunZoUKtrMrjzB(+W8<D{+u~EbN<D+}BtOX8
zB464ca%ny1VNMVWa<*zeK+n)t(sK<PakK8BEfd;a-?HEPQPKYSuY30I|0<{O$)jDD
z(~UK<Om6;V)%=5iJ$MG+TkU8z+mposfHVO>lg$8XQ*(O>8V}vrj{jt*-nSnVa6qjg
z{c*Azr@hw0Ud)3_`<X5c60i6mKA6P;faKEPre$myhS<>L;=rc42)@CIE&H<Y>VIm!
z`>;SY7YBU(XHVYj0NBgV6wbyv&O~x>v-Sf3S#ojUg=K6Ru(5&Q1vZh`rhV|kg8kb+
z%N6+fFN>0+;2{Ubx9lH2FID6p1mbc%i7T2Z6|4OKT|m-x^@Z)ax@k~$UU@-M?MnT1
zfi)F%)qJ!RO5a=SQ}N2$D;pUc;L2vwd1b>gU<e+>+bAQS4xl;?&T;P-7_QlLaj$RK
z4z3GgG6!&IOeFELu5-abvN!+`CLR4SOa-X9)3Ip>4jsmw)ZEY))Ts^i?Ud_zIPN(q
zUNKA5kB^7Ps{H`%{r&*$m)9Aj<FJ7Y5ym`BWD{pw?Dut1KK4c*NwGx_cv5`*862SP
z@_N$Nw{8iSMfPnN)<O13W6@o#B9{ktKH}rL*1mPA^*-cef5;Vb;<92>O~C8s&Ef#P
z>q%?xx<w#qYwsHA0dN5v*4rR2h$Lv7bJWjLeUao|m_z+kO8%Oc!5q{x=z+7x`yBeU
z_N*lrde*E%FG_4C()rfxN53w_f3oTQ=`SVQdu_wGe38SGFU<DKSO4!&?LB(DL%ndp
z7U5BunhA4yGB^N`CoR~dd(8k;a%0gk-aWJ)&zgEa_{p{v9tP;2J}$;-ul26wHmGMy
z_w&>YV*5Vu<Hh?N`Zag2B}mqMiOHLx3E|S3A&A{QxMr7E_TyRI8_KN{2k>zX=pTMk
z{fjy1VUH&5OOj`Dv$2nUESx>w=g<#Hnl3&!O)xM4^^2<_z3)6=aAL!J_^X1Aar^$J
zuYm!%hUNEvyj_uh5OVV`q~7%=i3fFaip=ByFnDsli-xx-&+{Un<I=1*cJVRDCw|RM
zg9k<_Hs`9YHNwNW!lg54zV-=q$`c-{U#<NBja{n=z$%cO^v+ci`n~VDak0*oApV?u
zuD&aZ#kJL9?4t!kPQ|&X_FPv62LRFp1_1JvSEFMUhm8EI27pJY%*R;Aio`e%{pnog
z1a!pp42pMXeb`U2Qu_hUb}oN#76v%mzG6r!$X>1D3-iJ_FJ0-E{FuK34@-wg`Ksi}
z994R?XK;Wstt%sE+L!H__7#L-iMB#*imQ%!<}BK6sH;zE+BimWsbSs6Q`}?+oRa7H
z0m2#f%H#kIZ7X%(Xu~0vEy`+0#afpQ>nnz#dMFhxhYlOZkrx*soetq*RFmckIh|em
z0q`{dQA6u8E@wfSh6@6+mr|e{h;gpifu#D1I^iX#t+^?g>Z83<ylFe516)|{su$w=
z3qUTYZ&@<rM_mpAaq?+iAmwnp0Q%E;pbH=8`aTg)P#@&r(vFjP>$5rlzWSEs<aEoD
zJ>9Y>STbHma8vI`Nk^hMN{I_X9$Y-E<@|uF!cO-N^F3dE%VKs1XlTwKJKena^i;D9
z0P<Axk^vh*Iu1f{va11dBp%eWO|={kYdE&#_*JjQi?b*%O8n8{J!*P*gi}rVr(bVc
zGzuI5Bn`s8yk)5Fr07sXI{UaTZ%;KXg?`7ah->3q#*#>xp*h1ZK;LmX!s1h1;QjNm
z97@vj<YO!v1m!wkTIc%E@AI7u0C~o9=P5Vlb+Rd+{Q>krlev@U^G{DU=FQ2*XVefS
z+urwnqk;z3xbvZnI{U?zI?Cf3<&OG<{n#JYo;bhwaXfdX?|4K07&PjFh6y!1_lyPs
zNxDcGSapu;LMPAZB0k3>ZxG-|uBuY>ebs#||NO++XASYpZ}pvQ{Lg#G&*ke*H0Hm0
z{M?r!^L3~zH0J&3cp4as<Dsuw?8~~ZH7$LmuHoss@q9<`=pDVIcl4t7UmvNh&FF?1
QEdT%j07*qoM6N<$f~id4od5s;

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_base/lark_base.py b/api/core/tools/provider/builtin/lark_base/lark_base.py
new file mode 100644
index 0000000000..de9b368311
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/lark_base.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkBaseProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_base/lark_base.yaml b/api/core/tools/provider/builtin/lark_base/lark_base.yaml
new file mode 100644
index 0000000000..200b2e22cf
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/lark_base.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_base
+  label:
+    en_US: Lark Base
+    zh_Hans: Lark 多维表格
+  description:
+    en_US: |
+      Lark base, requires the following permissions: bitable:app.
+    zh_Hans: |
+      Lark 多维表格，需要开通以下权限: bitable:app。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_base/tools/add_records.py b/api/core/tools/provider/builtin/lark_base/tools/add_records.py
new file mode 100644
index 0000000000..c46898062a
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/add_records.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class AddRecordsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_id = tool_parameters.get("table_id")
+        table_name = tool_parameters.get("table_name")
+        records = tool_parameters.get("records")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.add_records(app_token, table_id, table_name, records, user_id_type)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/add_records.yaml b/api/core/tools/provider/builtin/lark_base/tools/add_records.yaml
new file mode 100644
index 0000000000..f2a93490dc
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/add_records.yaml
@@ -0,0 +1,91 @@
+identity:
+  name: add_records
+  author: Doug Lea
+  label:
+    en_US: Add Records
+    zh_Hans: 新增多条记录
+description:
+  human:
+    en_US: Add Multiple Records to Multidimensional Table
+    zh_Hans: 在多维表格数据表中新增多条记录
+  llm: A tool for adding multiple records to a multidimensional table. (在多维表格数据表中新增多条记录)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_id
+    type: string
+    required: false
+    label:
+      en_US: table_id
+      zh_Hans: table_id
+    human_description:
+      en_US: Unique identifier for the multidimensional table data, either table_id or table_name must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: false
+    label:
+      en_US: table_name
+      zh_Hans: table_name
+    human_description:
+      en_US: Name of the multidimensional table data, either table_name or table_id must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: records
+    type: string
+    required: true
+    label:
+      en_US: records
+      zh_Hans: 记录列表
+    human_description:
+      en_US: |
+        List of records to be added in this request. Example value: [{"multi-line-text":"text content","single_select":"option 1","date":1674206443000}]
+        For supported field types, refer to the integration guide (https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification). For data structures of different field types, refer to the data structure overview (https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure).
+      zh_Hans: |
+        本次请求将要新增的记录列表，示例值：[{"多行文本":"文本内容","单选":"选项 1","日期":1674206443000}]。
+        当前接口支持的字段类型请参考接入指南(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification)，不同类型字段的数据结构请参考数据结构概述(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure)。
+    llm_description: |
+      本次请求将要新增的记录列表，示例值：[{"多行文本":"文本内容","单选":"选项 1","日期":1674206443000}]。
+      当前接口支持的字段类型请参考接入指南(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification)，不同类型字段的数据结构请参考数据结构概述(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure)。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_base/tools/create_base.py b/api/core/tools/provider/builtin/lark_base/tools/create_base.py
new file mode 100644
index 0000000000..a857c6ced6
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/create_base.py
@@ -0,0 +1,18 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateBaseTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        name = tool_parameters.get("name")
+        folder_token = tool_parameters.get("folder_token")
+
+        res = client.create_base(name, folder_token)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/create_base.yaml b/api/core/tools/provider/builtin/lark_base/tools/create_base.yaml
new file mode 100644
index 0000000000..e622edf336
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/create_base.yaml
@@ -0,0 +1,42 @@
+identity:
+  name: create_base
+  author: Doug Lea
+  label:
+    en_US: Create Base
+    zh_Hans: 创建多维表格
+description:
+  human:
+    en_US: Create Multidimensional Table in Specified Directory
+    zh_Hans: 在指定目录下创建多维表格
+  llm: A tool for creating a multidimensional table in a specified directory. (在指定目录下创建多维表格)
+parameters:
+  - name: name
+    type: string
+    required: false
+    label:
+      en_US: name
+      zh_Hans: 多维表格 App 名字
+    human_description:
+      en_US: |
+        Name of the multidimensional table App. Example value: "A new multidimensional table".
+      zh_Hans: 多维表格 App 名字，示例值："一篇新的多维表格"。
+    llm_description: 多维表格 App 名字，示例值："一篇新的多维表格"。
+    form: llm
+
+  - name: folder_token
+    type: string
+    required: false
+    label:
+      en_US: folder_token
+      zh_Hans: 多维表格 App 归属文件夹
+    human_description:
+      en_US: |
+        Folder where the multidimensional table App belongs. Default is empty, meaning the table will be created in the root directory of the cloud space. Example values: Lf8uf6BoAlWkUfdGtpMjUV0PpZd or https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd.
+        The folder_token must be an existing folder and supports inputting folder token or folder URL.
+      zh_Hans: |
+        多维表格 App 归属文件夹。默认为空，表示多维表格将被创建在云空间根目录。示例值: Lf8uf6BoAlWkUfdGtpMjUV0PpZd 或者 https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd。
+        folder_token 必须是已存在的文件夹，支持输入文件夹 token 或者文件夹 URL。
+    llm_description: |
+      多维表格 App 归属文件夹。默认为空，表示多维表格将被创建在云空间根目录。示例值: Lf8uf6BoAlWkUfdGtpMjUV0PpZd 或者 https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd。
+      folder_token 必须是已存在的文件夹，支持输入文件夹 token 或者文件夹 URL。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/create_table.py b/api/core/tools/provider/builtin/lark_base/tools/create_table.py
new file mode 100644
index 0000000000..aff7e715b7
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/create_table.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateTableTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_name = tool_parameters.get("table_name")
+        default_view_name = tool_parameters.get("default_view_name")
+        fields = tool_parameters.get("fields")
+
+        res = client.create_table(app_token, table_name, default_view_name, fields)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/create_table.yaml b/api/core/tools/provider/builtin/lark_base/tools/create_table.yaml
new file mode 100644
index 0000000000..8b1007b9a5
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/create_table.yaml
@@ -0,0 +1,61 @@
+identity:
+  name: create_table
+  author: Doug Lea
+  label:
+    en_US: Create Table
+    zh_Hans: 新增数据表
+description:
+  human:
+    en_US: Add a Data Table to Multidimensional Table
+    zh_Hans: 在多维表格中新增一个数据表
+  llm: A tool for adding a data table to a multidimensional table. (在多维表格中新增一个数据表)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: true
+    label:
+      en_US: Table Name
+      zh_Hans: 数据表名称
+    human_description:
+      en_US: |
+        The name of the data table, length range: 1 character to 100 characters.
+      zh_Hans: 数据表名称，长度范围：1 字符 ～ 100 字符。
+    llm_description: 数据表名称，长度范围：1 字符 ～ 100 字符。
+    form: llm
+
+  - name: default_view_name
+    type: string
+    required: false
+    label:
+      en_US: Default View Name
+      zh_Hans: 默认表格视图的名称
+    human_description:
+      en_US: The name of the default table view, defaults to "Table" if not filled.
+      zh_Hans: 默认表格视图的名称，不填则默认为"表格"。
+    llm_description: 默认表格视图的名称，不填则默认为"表格"。
+    form: llm
+
+  - name: fields
+    type: string
+    required: true
+    label:
+      en_US: Initial Fields
+      zh_Hans: 初始字段
+    human_description:
+      en_US: |
+        Initial fields of the data table, format: [ { "field_name": "Multi-line Text","type": 1 },{ "field_name": "Number","type": 2 },{ "field_name": "Single Select","type": 3 },{ "field_name": "Multiple Select","type": 4 },{ "field_name": "Date","type": 5 } ]. For field details, refer to: https://open.larkoffice.com/document/server-docs/docs/bitable-v1/app-table-field/guide
+      zh_Hans: 数据表的初始字段，格式为：[{"field_name":"多行文本","type":1},{"field_name":"数字","type":2},{"field_name":"单选","type":3},{"field_name":"多选","type":4},{"field_name":"日期","type":5}]。字段详情参考：https://open.larkoffice.com/document/server-docs/docs/bitable-v1/app-table-field/guide
+    llm_description: 数据表的初始字段，格式为：[{"field_name":"多行文本","type":1},{"field_name":"数字","type":2},{"field_name":"单选","type":3},{"field_name":"多选","type":4},{"field_name":"日期","type":5}]。字段详情参考：https://open.larkoffice.com/document/server-docs/docs/bitable-v1/app-table-field/guide
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/delete_records.py b/api/core/tools/provider/builtin/lark_base/tools/delete_records.py
new file mode 100644
index 0000000000..1b0a747050
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/delete_records.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class DeleteRecordsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_id = tool_parameters.get("table_id")
+        table_name = tool_parameters.get("table_name")
+        record_ids = tool_parameters.get("record_ids")
+
+        res = client.delete_records(app_token, table_id, table_name, record_ids)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/delete_records.yaml b/api/core/tools/provider/builtin/lark_base/tools/delete_records.yaml
new file mode 100644
index 0000000000..c30ebd630c
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/delete_records.yaml
@@ -0,0 +1,86 @@
+identity:
+  name: delete_records
+  author: Doug Lea
+  label:
+    en_US: Delete Records
+    zh_Hans: 删除多条记录
+description:
+  human:
+    en_US: Delete Multiple Records from Multidimensional Table
+    zh_Hans: 删除多维表格数据表中的多条记录
+  llm: A tool for deleting multiple records from a multidimensional table. (删除多维表格数据表中的多条记录)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_id
+    type: string
+    required: false
+    label:
+      en_US: table_id
+      zh_Hans: table_id
+    human_description:
+      en_US: Unique identifier for the multidimensional table data, either table_id or table_name must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: false
+    label:
+      en_US: table_name
+      zh_Hans: table_name
+    human_description:
+      en_US: Name of the multidimensional table data, either table_name or table_id must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: record_ids
+    type: string
+    required: true
+    label:
+      en_US: Record IDs
+      zh_Hans: 记录 ID 列表
+    human_description:
+      en_US: |
+        List of IDs for the records to be deleted, example value: ["recwNXzPQv"].
+      zh_Hans: 删除的多条记录 ID 列表，示例值：["recwNXzPQv"]。
+    llm_description: 删除的多条记录 ID 列表，示例值：["recwNXzPQv"]。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_base/tools/delete_tables.py b/api/core/tools/provider/builtin/lark_base/tools/delete_tables.py
new file mode 100644
index 0000000000..e0ecae2f17
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/delete_tables.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class DeleteTablesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_ids = tool_parameters.get("table_ids")
+        table_names = tool_parameters.get("table_names")
+
+        res = client.delete_tables(app_token, table_ids, table_names)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/delete_tables.yaml b/api/core/tools/provider/builtin/lark_base/tools/delete_tables.yaml
new file mode 100644
index 0000000000..498126eae5
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/delete_tables.yaml
@@ -0,0 +1,49 @@
+identity:
+  name: delete_tables
+  author: Doug Lea
+  label:
+    en_US: Delete Tables
+    zh_Hans: 删除数据表
+description:
+  human:
+    en_US: Batch Delete Data Tables from Multidimensional Table
+    zh_Hans: 批量删除多维表格中的数据表
+  llm: A tool for batch deleting data tables from a multidimensional table. (批量删除多维表格中的数据表)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_ids
+    type: string
+    required: false
+    label:
+      en_US: Table IDs
+      zh_Hans: 数据表 ID
+    human_description:
+      en_US: |
+        IDs of the tables to be deleted. Each operation supports deleting up to 50 tables. Example: ["tbl1TkhyTWDkSoZ3"]. Ensure that either table_ids or table_names is not empty.
+      zh_Hans: 待删除的数据表的 ID，每次操作最多支持删除 50 个数据表。示例值：["tbl1TkhyTWDkSoZ3"]。请确保 table_ids 和 table_names 至少有一个不为空。
+    llm_description: 待删除的数据表的 ID，每次操作最多支持删除 50 个数据表。示例值：["tbl1TkhyTWDkSoZ3"]。请确保 table_ids 和 table_names 至少有一个不为空。
+    form: llm
+
+  - name: table_names
+    type: string
+    required: false
+    label:
+      en_US: Table Names
+      zh_Hans: 数据表名称
+    human_description:
+      en_US: |
+        Names of the tables to be deleted. Each operation supports deleting up to 50 tables. Example: ["Table1", "Table2"]. Ensure that either table_names or table_ids is not empty.
+      zh_Hans: 待删除的数据表的名称，每次操作最多支持删除 50 个数据表。示例值：["数据表1", "数据表2"]。请确保 table_names 和 table_ids 至少有一个不为空。
+    llm_description: 待删除的数据表的名称，每次操作最多支持删除 50 个数据表。示例值：["数据表1", "数据表2"]。请确保 table_names 和 table_ids 至少有一个不为空。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/get_base_info.py b/api/core/tools/provider/builtin/lark_base/tools/get_base_info.py
new file mode 100644
index 0000000000..2c23248b88
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/get_base_info.py
@@ -0,0 +1,17 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetBaseInfoTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+
+        res = client.get_base_info(app_token)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/get_base_info.yaml b/api/core/tools/provider/builtin/lark_base/tools/get_base_info.yaml
new file mode 100644
index 0000000000..eb0e7a26c0
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/get_base_info.yaml
@@ -0,0 +1,23 @@
+identity:
+  name: get_base_info
+  author: Doug Lea
+  label:
+    en_US: Get Base Info
+    zh_Hans: 获取多维表格元数据
+description:
+  human:
+    en_US: Get Metadata Information of Specified Multidimensional Table
+    zh_Hans: 获取指定多维表格的元数据信息
+  llm: A tool for getting metadata information of a specified multidimensional table. (获取指定多维表格的元数据信息)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/list_tables.py b/api/core/tools/provider/builtin/lark_base/tools/list_tables.py
new file mode 100644
index 0000000000..55b706854b
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/list_tables.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ListTablesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        page_token = tool_parameters.get("page_token")
+        page_size = tool_parameters.get("page_size", 20)
+
+        res = client.list_tables(app_token, page_token, page_size)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/list_tables.yaml b/api/core/tools/provider/builtin/lark_base/tools/list_tables.yaml
new file mode 100644
index 0000000000..7571519039
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/list_tables.yaml
@@ -0,0 +1,50 @@
+identity:
+  name: list_tables
+  author: Doug Lea
+  label:
+    en_US: List Tables
+    zh_Hans: 列出数据表
+description:
+  human:
+    en_US: Get All Data Tables under Multidimensional Table
+    zh_Hans: 获取多维表格下的所有数据表
+  llm: A tool for getting all data tables under a multidimensional table. (获取多维表格下的所有数据表)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: page_size
+    type: number
+    required: false
+    default: 20
+    label:
+      en_US: page_size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: |
+        Page size, default value: 20, maximum value: 100.
+      zh_Hans: 分页大小，默认值：20，最大值：100。
+    llm_description: 分页大小，默认值：20，最大值：100。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: page_token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: |
+        Page token, leave empty for the first request to start from the beginning; a new page_token will be returned if there are more items in the paginated query results, which can be used for the next traversal. Example value: "tblsRc9GRRXKqhvW".
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。示例值："tblsRc9GRRXKqhvW"。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。示例值："tblsRc9GRRXKqhvW"。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/read_records.py b/api/core/tools/provider/builtin/lark_base/tools/read_records.py
new file mode 100644
index 0000000000..5cf25aad84
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/read_records.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ReadRecordsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_id = tool_parameters.get("table_id")
+        table_name = tool_parameters.get("table_name")
+        record_ids = tool_parameters.get("record_ids")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.read_records(app_token, table_id, table_name, record_ids, user_id_type)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/read_records.yaml b/api/core/tools/provider/builtin/lark_base/tools/read_records.yaml
new file mode 100644
index 0000000000..911e667cfc
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/read_records.yaml
@@ -0,0 +1,86 @@
+identity:
+  name: read_records
+  author: Doug Lea
+  label:
+    en_US: Read Records
+    zh_Hans: 批量获取记录
+description:
+  human:
+    en_US: Batch Retrieve Records from Multidimensional Table
+    zh_Hans: 批量获取多维表格数据表中的记录信息
+  llm: A tool for batch retrieving records from a multidimensional table, supporting up to 100 records per call. (批量获取多维表格数据表中的记录信息，单次调用最多支持查询 100 条记录)
+
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_id
+    type: string
+    required: false
+    label:
+      en_US: table_id
+      zh_Hans: table_id
+    human_description:
+      en_US: Unique identifier for the multidimensional table data, either table_id or table_name must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: false
+    label:
+      en_US: table_name
+      zh_Hans: table_name
+    human_description:
+      en_US: Name of the multidimensional table data, either table_name or table_id must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: record_ids
+    type: string
+    required: true
+    label:
+      en_US: record_ids
+      zh_Hans: 记录 ID 列表
+    human_description:
+      en_US: List of record IDs, which can be obtained by calling the "Query Records API".
+      zh_Hans: 记录 ID 列表，可以通过调用"查询记录接口"获取。
+    llm_description: 记录 ID 列表，可以通过调用"查询记录接口"获取。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_base/tools/search_records.py b/api/core/tools/provider/builtin/lark_base/tools/search_records.py
new file mode 100644
index 0000000000..9b0abcf067
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/search_records.py
@@ -0,0 +1,39 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class SearchRecordsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_id = tool_parameters.get("table_id")
+        table_name = tool_parameters.get("table_name")
+        view_id = tool_parameters.get("view_id")
+        field_names = tool_parameters.get("field_names")
+        sort = tool_parameters.get("sort")
+        filters = tool_parameters.get("filter")
+        page_token = tool_parameters.get("page_token")
+        automatic_fields = tool_parameters.get("automatic_fields", False)
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+        page_size = tool_parameters.get("page_size", 20)
+
+        res = client.search_record(
+            app_token,
+            table_id,
+            table_name,
+            view_id,
+            field_names,
+            sort,
+            filters,
+            page_token,
+            automatic_fields,
+            user_id_type,
+            page_size,
+        )
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/search_records.yaml b/api/core/tools/provider/builtin/lark_base/tools/search_records.yaml
new file mode 100644
index 0000000000..edd86ab9d6
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/search_records.yaml
@@ -0,0 +1,163 @@
+identity:
+  name: search_records
+  author: Doug Lea
+  label:
+    en_US: Search Records
+    zh_Hans: 查询记录
+description:
+  human:
+    en_US: Query records in a multidimensional table, up to 500 rows per query.
+    zh_Hans: 查询多维表格数据表中的记录，单次最多查询 500 行记录。
+  llm: A tool for querying records in a multidimensional table, up to 500 rows per query. (查询多维表格数据表中的记录，单次最多查询 500 行记录)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_id
+    type: string
+    required: false
+    label:
+      en_US: table_id
+      zh_Hans: table_id
+    human_description:
+      en_US: Unique identifier for the multidimensional table data, either table_id or table_name must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: false
+    label:
+      en_US: table_name
+      zh_Hans: table_name
+    human_description:
+      en_US: Name of the multidimensional table data, either table_name or table_id must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: view_id
+    type: string
+    required: false
+    label:
+      en_US: view_id
+      zh_Hans: 视图唯一标识
+    human_description:
+      en_US: |
+        Unique identifier for a view in a multidimensional table. It can be found in the URL's query parameter with the key 'view'. For example: https://lark-japan.jp.larksuite.com/base/XXX0bfYEraW5OWsbhcFjEqj6pxh?table=tbl5I6jqwz8wBRMv&view=vewW5zXVEU.
+      zh_Hans: 多维表格中视图的唯一标识，可在多维表格的 URL 地址栏中找到，query 参数中 key 为 view 的部分。例如：https://lark-japan.jp.larksuite.com/base/XXX0bfYEraW5OWsbhcFjEqj6pxh?table=tbl5I6jqwz8wBRMv&view=vewW5zXVEU。
+    llm_description: 多维表格中视图的唯一标识，可在多维表格的 URL 地址栏中找到，query 参数中 key 为 view 的部分。例如：https://lark-japan.jp.larksuite.com/base/XXX0bfYEraW5OWsbhcFjEqj6pxh?table=tbl5I6jqwz8wBRMv&view=vewW5zXVEU。
+    form: llm
+
+  - name: field_names
+    type: string
+    required: false
+    label:
+      en_US: field_names
+      zh_Hans: 字段名称
+    human_description:
+      en_US: |
+        Field names to specify which fields to include in the returned records. Example value: ["Field1", "Field2"].
+      zh_Hans: 字段名称，用于指定本次查询返回记录中包含的字段。示例值：["字段1","字段2"]。
+    llm_description: 字段名称，用于指定本次查询返回记录中包含的字段。示例值：["字段1","字段2"]。
+    form: llm
+
+  - name: sort
+    type: string
+    required: false
+    label:
+      en_US: sort
+      zh_Hans: 排序条件
+    human_description:
+      en_US: |
+        Sorting conditions, for example: [{"field_name":"Multiline Text","desc":true}].
+      zh_Hans: 排序条件，例如：[{"field_name":"多行文本","desc":true}]。
+    llm_description: 排序条件，例如：[{"field_name":"多行文本","desc":true}]。
+    form: llm
+
+  - name: filter
+    type: string
+    required: false
+    label:
+      en_US: filter
+      zh_Hans: 筛选条件
+    human_description:
+      en_US: Object containing filter information. For details on how to fill in the filter, refer to the record filter parameter guide (https://open.larkoffice.com/document/uAjLw4CM/ukTMukTMukTM/reference/bitable-v1/app-table-record/record-filter-guide).
+      zh_Hans: 包含条件筛选信息的对象。了解如何填写 filter，参考记录筛选参数填写指南(https://open.larkoffice.com/document/uAjLw4CM/ukTMukTMukTM/reference/bitable-v1/app-table-record/record-filter-guide)。
+    llm_description: 包含条件筛选信息的对象。了解如何填写 filter，参考记录筛选参数填写指南(https://open.larkoffice.com/document/uAjLw4CM/ukTMukTMukTM/reference/bitable-v1/app-table-record/record-filter-guide)。
+    form: llm
+
+  - name: automatic_fields
+    type: boolean
+    required: false
+    label:
+      en_US: automatic_fields
+      zh_Hans: automatic_fields
+    human_description:
+      en_US: Whether to return automatically calculated fields. Default is false, meaning they are not returned.
+      zh_Hans: 是否返回自动计算的字段。默认为 false，表示不返回。
+    llm_description: 是否返回自动计算的字段。默认为 false，表示不返回。
+    form: form
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: page_size
+    type: number
+    required: false
+    default: 20
+    label:
+      en_US: page_size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: |
+        Page size, default value: 20, maximum value: 500.
+      zh_Hans: 分页大小，默认值：20，最大值：500。
+    llm_description: 分页大小，默认值：20，最大值：500。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: page_token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: |
+        Page token, leave empty for the first request to start from the beginning; a new page_token will be returned if there are more items in the paginated query results, which can be used for the next traversal. Example value: "tblsRc9GRRXKqhvW".
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。示例值："tblsRc9GRRXKqhvW"。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。示例值："tblsRc9GRRXKqhvW"。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_base/tools/update_records.py b/api/core/tools/provider/builtin/lark_base/tools/update_records.py
new file mode 100644
index 0000000000..7c263df2bb
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/update_records.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class UpdateRecordsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        app_token = tool_parameters.get("app_token")
+        table_id = tool_parameters.get("table_id")
+        table_name = tool_parameters.get("table_name")
+        records = tool_parameters.get("records")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.update_records(app_token, table_id, table_name, records, user_id_type)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_base/tools/update_records.yaml b/api/core/tools/provider/builtin/lark_base/tools/update_records.yaml
new file mode 100644
index 0000000000..68117e7136
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_base/tools/update_records.yaml
@@ -0,0 +1,91 @@
+identity:
+  name: update_records
+  author: Doug Lea
+  label:
+    en_US: Update Records
+    zh_Hans: 更新多条记录
+description:
+  human:
+    en_US: Update Multiple Records in Multidimensional Table
+    zh_Hans: 更新多维表格数据表中的多条记录
+  llm: A tool for updating multiple records in a multidimensional table. (更新多维表格数据表中的多条记录)
+parameters:
+  - name: app_token
+    type: string
+    required: true
+    label:
+      en_US: app_token
+      zh_Hans: app_token
+    human_description:
+      en_US: Unique identifier for the multidimensional table, supports inputting document URL.
+      zh_Hans: 多维表格的唯一标识符，支持输入文档 URL。
+    llm_description: 多维表格的唯一标识符，支持输入文档 URL。
+    form: llm
+
+  - name: table_id
+    type: string
+    required: false
+    label:
+      en_US: table_id
+      zh_Hans: table_id
+    human_description:
+      en_US: Unique identifier for the multidimensional table data, either table_id or table_name must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的唯一标识符，table_id 和 table_name 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: table_name
+    type: string
+    required: false
+    label:
+      en_US: table_name
+      zh_Hans: table_name
+    human_description:
+      en_US: Name of the multidimensional table data, either table_name or table_id must be provided, cannot be empty simultaneously.
+      zh_Hans: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    llm_description: 多维表格数据表的名称，table_name 和 table_id 至少需要提供一个，不能同时为空。
+    form: llm
+
+  - name: records
+    type: string
+    required: true
+    label:
+      en_US: records
+      zh_Hans: 记录列表
+    human_description:
+      en_US: |
+        List of records to be updated in this request. Example value: [{"fields":{"multi-line-text":"text content","single_select":"option 1","date":1674206443000},"record_id":"recupK4f4RM5RX"}].
+        For supported field types, refer to the integration guide (https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification). For data structures of different field types, refer to the data structure overview (https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure).
+      zh_Hans: |
+        本次请求将要更新的记录列表，示例值：[{"fields":{"多行文本":"文本内容","单选":"选项 1","日期":1674206443000},"record_id":"recupK4f4RM5RX"}]。
+        当前接口支持的字段类型请参考接入指南(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification)，不同类型字段的数据结构请参考数据结构概述(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure)。
+    llm_description: |
+      本次请求将要更新的记录列表，示例值：[{"fields":{"多行文本":"文本内容","单选":"选项 1","日期":1674206443000},"record_id":"recupK4f4RM5RX"}]。
+      当前接口支持的字段类型请参考接入指南(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/notification)，不同类型字段的数据结构请参考数据结构概述(https://open.larkoffice.com/document/server-docs/docs/bitable-v1/bitable-structure)。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_calendar/_assets/icon.png b/api/core/tools/provider/builtin/lark_calendar/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..2a934747a98c6680065941bcd31d2400da1eaf23
GIT binary patch
literal 5488
zcmd^@S6CClyT+3cB}hx?J%AJ?fE0mHMQMV8&_R?YO?nelY7hlNk>0Cxkg60F3{fc}
zH9!DCT0jJ80tr>&<bSTu?KwB+Vt1Z>_S<*on|*iYn;+KLNRROn&m{l=z^H#)+mzb=
z`Og8lNIh#<tfx^MV4$hqEkMIC-#P%mE}^e|)BK6uR-s`ETVH0Mz2)PP*iM*g8O5=u
z=gVzzzUD!>4qnG3;S|irW=|=<j2DR<b4O5KP?)3Ik%Q1@$SCYC=t~jV@og5Go&l+;
zQFk;0Sx(5&rKN{k<|!i`a<&`VhktVq$&Aa}S{%RWkl(k5UE;6YBZhWmN7`8iyHpC_
z7X*NYS&WB6nEGx4zz~qGH~<(+r-^40XKsiFfD<(UASPA}hTllwJ}nT!6%GLNi2;DB
z1#0h@Kv0<z00fBwfP#@v{rJIx4rCzIF9!hL^xIohfoS5_XapB!fRO2>l}|xX42D3f
ziFX7+o4)i%ZwLZ_Wy~1N7x0X<1J061*y29gfDyJ2fUTeVMQPHq7@KLU>F~{ov1emn
zTTR^lX{pNZoq%B_m`)+II}6z;t*n3|Mo<iR8#eBehwYo7A3tR9cJY}v1@1o?0R+81
z#7^wzs_W1|*B*gTRrjP;vLHEqrzh1NawpkKee*^mz`ry=Ef6Zccj8bWk$Xm;_~Qd+
z!$Rpq6vkMw2T_W`z`@_hSLPi;t-USqYIZ!O+oWq)HnGaTYZ-CEyis!gE(t?eX-|+*
z_pAj^pNK_U;<C>U1jS;dIL*?`4jeJEOe)(8nc{{SYk($W#)H9s-F>-9L)q>vds%qH
zkcrltA7vCoxnI4*96v@UvWzE-7Cx7P{&@x%&*6D0Sj%Vv92rJtf5apSYHAh%#><`<
z2e{f;;;o92hA{8!WL}P&cvjHhvd7BO#hQoR;KbjEOqS{00syPkr2de>038=`YCj_O
zPd5aDqo-cUk32YjA+$q>s&#{_U4>P|VUB0oV{Y#n(#1N^fI)(I^SU~-b7vfzi7wtM
z^w6|?o(Z6K3k;+Kp}J2e-ZCfW^Q6Q6wEFUk_k^MfZ-zo4F__Xr;-SE%Op=u@zWCtx
zh*)#))OFs6a3<iYeNAfYxhqdoBBoK@Kb<X$K2MBhD_YptazumcdXlv+J|{|Xh-M~M
zk7}-Kzv8AZQ(d9guYF|+MzzF%vhYD1)kG$8FoP#x!NorTFXPmO0l?4<)N>XAAOzW2
zaQQ(p)m{RktgoK`-$<zF;pO`bTxxeUpD1f~3#O!N4*qPQX&s)<+#`ItT};OS{52?=
zci}ds=0~_l42A}!e(NBLnu8VxOsk2SgM(QzKAh_L(-fU<s3bLq>50pJaB2=3&M;Z<
z9qKSBQXoHQsl&j0SDWDz3kDNys17>(|0S{QY(fc-ee#?BF<;wCO^An-3EEzIBkWv}
zKsIk^<Npo;Z8x=$$tEqUr{$_OAps?4=Vwu(Mtdvjr2cDZ!(8b5VrP4b{>KKf7o0v7
zo)OJwp1u-Fzm8t+m$NYP78%<NdndTa^tF^@>(2mN{nOOaPn7rk52RKXip9_VJ&iy2
z(`YX=^0A&q8oV%!9lT#qgwn1%Cj_27!_4M<F4B-l+<dk6ey%)#i7tEJ&T4Ju#TT{(
z<ajUO1>D;@oHK;g{)Qou*`7VPiE&kMQBhO8^r17_RKNU12a>1wxuj`0t0s!bY)4s!
zD?#4%fVSZ!q_)Gvm~T(j8f<T9wI(=i<4)-;?aw}-GRsC*K21OM7F^_5btc+Kd?fpC
z({Y)hgA^KyW=Sg@@Y)@@@39y|m(+&v79Umiv!6K>p+Ga=%Bt#<jqsO@aQ4iDNa;Q1
zVk=_%nWFKw1`D&|P-r@DMh$T8jPR)%rF|z-K2>I3m-oc&6^U8Hr;PRv-1eVCti15c
z82eYGwqp7}`)hNLUtR=Dj$X5rBizuen46llf69Agt=t;*q;xz|RhMGh9D57TaiOlK
zW54w4=tz9_xY_l<09!q@*Q*o7el@U{4iA>_x$Ur8rgLtd1}wROt^d})#PqKdN^EzJ
z7h0^03z&m4`l2JG1D0xSRrOoZb~{KblW?<gFvtjabwf%$oN_pSaY4pSfmnNr*s0rS
zD$X_zH~T37GJ0`!+%TZ)u4_mgVQ9CTy|!q^!CC8#Y^|?B9@NjknA-=j)wN+xDWtCt
zbgiIDK30Mt6>80uK9ILNkR>xjR`b&#oVou@Cmm6V#ukF4sbg!;3bqq=X%cB7Tqeui
z9NUQemDNQ#3o>0InrU5C`{tK$c6a^jxOY*iE@dc1HVm6dXZbe&l(yRQe9rZ8#aZZ^
zw6z&J(neQ_c6Wlb6zOxXR{3k(gvGzbShj_u7a8vdgN?)Gg|#jWL%d$hipGw#^159)
zqu^pbdyA};j?!M<rO2EA^NZ!nKxI(_m&TN|@VTi{vn5AkdTil`I!@6xovqAou(z$2
zOY!3Q{jvPM>fU@>0yNb)<?5Sw<r1lJ?FR?SclRC#AT55{Zt6NZE3qqfdGn19PA{}9
zsdEvPtVn{3_ouTIf3KG{5E>JcavNAAk|JC>{Yv@aK_w@vn<`L0BfkXk49mu<G0*J8
zM12N(`ut?kQ?K0n^~GOrQHBbyL6Axx`gm#;7OK*g^6as6B6K%AID4yucbgb;*et1E
zN7H7T12PSme^yEO>LbB}i?kf}5yCmO@z)g%j_r`dtUNCy{`O2m4Bp};<onh|CQEto
zD7+Ed-Cobbn1m;NV2IF_>pnCLqyyd=Jo}x_XVBrx^@8}}67`{sD|1|051pQ4fgjz{
z37jS;^1aQzysFD%sxq~W{vU)KfX}hm6@2}C7_E=!#mXw}$z_AVTN^$9LQ=ct=!sU(
zuSqmPdWU6%TT(R<S?r}Jmx8z<)(~gfoYa}ynk3sW9X<n-b;iwo$uSo}2Uy~k(i@Ms
z55cxd2U;YjG*=_-^FNo^+YPq%>kiv}7Jk>bHrt%5g1z{g|ItE<ctv()^<UTes>vD=
z&CY);`g;{Wtx^Mmwlh#Ghh;4r7i%m$_j#>HO{r3ue5tQ-+PnW|wV65V(kEG?>=#a5
zZ^@jEZ+%y9ROh%3mK@ueb=Cg;(Tdg~{7CX>JD=m-e2)k&u!ziJnD<xscxmD_Jt$bI
zknq(`!sjg48qmi4(K32OT->5%qZZyAg8W|eHJ=)BJ@en|8qf1zrNdV9JR(N58HEGA
z5spvQ>TQu~9oAo>J*ex}I>_7SGSXdv6HEDxX!>EArh?9@3GhfiC31nXHaY5>MU~w%
zJ5P<!laGOPB`4B@DIETkABv=FzO7>ZibWvcY>CH6<w`lRAbavgA=`YtS;jV5ggW7}
z9Vc2$G`92#F1D1<L>80FGhbf8?IUW7<~$VM8QrKLtS?;$d$sO-<8GI6&twn^qBJ8P
zRT?$m=--sZ5<akO=!hEGLYrvT-^jQV;}7kG6+(6Q2ADL(Jmyp$-sFQe38L;)mW$W7
z<WDYyQnW%&Yg*0&mu=;w9K9W~+;w-b-htY}{0v}8L-Wa^O#RmhPUVsGt>OJr+$~${
zo&TPY0`_c{1}^{~`_y`&Y(#fLb{N_PA7+LHiH48nhxfhLK&boooRx_b0f2XmTO7U+
z^<OKqp&Ke2l$^WR-fF&orT=g15oI&WQhj(+8QR2s15q!9vRO|)qU7pXx)pHZ<8%*<
z{r6=NPwIFJi^_8^kGzw;%m}=<!;}kO+$<+5rAkPwM^BP993OCs(__i;SM&006H7*J
zZ^K7b)3E9lK5;b%{Wy+Ida*@Cuq41_^!Y%#<$?oLKx6-le*6vfo6Q>HPYDdWf4Tcl
zsw9+!RONFo0O@N-Ml)v9E%!g&DM}xV?9?Hi{}_9;)_7uN=Km`){jBTj&u=n3AP)WZ
z_pmAPdJbrcinuatr-OGK-a0dmu9%We3Gw)<1`iotEL<8s*QJJjZLOhW7OUqFPC=3~
zx}>Li@^{PPLqu^hr#@b3T$Lv#jr=%A0kBt^b8g!<+GZ#)1v#UG7S4^+J3pvXVwU;n
zVU=vQrF8D-(d!cmnU-E~ovcUM*s2<}qG!f;MP3FIi0uL7T-P<rqmDs6kv1h|8HgY5
zDp>NFY&OiE`r?be5iJRKDTVtwq<2%6lC6V8e-6cGm?YS${34;8=(N+NdW}0m<Ldq=
zYy{^d{Z&-M32r<eg_`!I2hO^6uF$oK)vGv6n!~+4$iTiY`gzJ;5e8*vhRC~gnPF?G
zJI^rk0ilm%Bv?%%NZiF!4YT2SEjEKq8&wLI>jFw;&v{V8oR>uqO1c-?Anj4^3)C*f
zT#==b66e{+N-En+9>_wbg@5Q&h8MJipzxr3eZ7aiqN>h$_8<^0BCi&X^Vp6zq=z;c
zexUaXRdJ`}V5T~6SHQ!Z!Pmb#Y)hpVr?|o!ll_&+k-8mGi*DtKHRJ>S>1BO{J8&Zz
z>Dr6)#$})#Ze9R00L8&!MJ*v=kbi2KcJk_{w_UFm9|vowb@b1n<qk+33onIQY!@}q
zHd10$k#DYUb$K#OJh@sdpmE-@9;|`Ne0vYOQ=s&+ua>$i6R<<1Tbk~~DSqDm?*#!Y
znZI?X0(UB$X(t!Ti4K-o$1D)@KR?Ih3x9j81xU9{!mnNO+S~0e7;=`|Sh>Msh<06W
zY=1LH$!%s{U2t*>R6U%7`uRK>P->BL*D-(UX4}_@mhuTrCO?b7zShW?ES9TOyJ&t@
zqLVDkKKb|c?(*`8kr}0o>6}hqj}FNDD+CLqc1dbx;Cn{QiVvUu`NZsE(d^6$m$43c
z1;E~ZRzNM;DuqO9s|)6Ch5hORd)-w80tY~CvhF?h^FHerJ$H+JSi+yG!nv9c7mxck
zk~Yb@WdGxDV`i)&r`3Np1>%clUC8-6PM_k#lM9(0D=ndZV)qgAUQS~M!yV)-)CgoO
zVcpw%O<;p@iUM?Iu1c#9oRH`*yta{8z^&l~Da4s6eH>3UJId>QHDdTdMz=NLIN2`M
zIw8LojX0Ds`}uI=U_|jHkbRoV@j*$`E|gs{=^K~lTXr*`85*9+^dpmZtSu$hjL6Au
z-QcE)Pk#F-E!LCB35vn+!ekooGu87X3g(_(LT_=Q|8iB?3;i!p8{KXr%L12MoUo>d
zdamz=FNS;v0Bv5PJBn>ta~?hdAfMqYL;O&`%Lc2S0pm?dxY*oUxJnc4qcTk)hh?%8
zi|2*s{kj_+=dtGKE{2Ex2vqB9ah1sPUq4AP7?%*R#GgGr(S!J`{xMTT*y93;o#3P4
zbi@T_;J_nB$P<^B)MEH{WW}UD=lQOf#ll?1*=B`NK*TU7#99mi^$UC%1cS4vUu-!x
z<LyH3%?+5Vw-0&m{vH)v46}9cM?jGLeJ6wD<^ya_uE!K##4p=KZjjN`Zr<Dlpwi{*
z=Z|;ahMB47>BHIV)Y~(kklJf(kp8S-$#gLY@^T-w>Q-y-P|*7Qy_VPhFZNW+)qF9L
zU+U-9f2zC$pYe{z4}%GR2Hr-R95YC{S0J|8cHP{n4?5CwN4PlxKI6E}9k_dgfE#o?
zhu1zHOgbx=3V>|BhR{85bhdhfYKm}y?Qb^lqq<qMYYt-fyQfA@to%))_8&E_dZ&e#
zoOF%<)#cpI`TLA4d=hLM_~7MPuvCpjq+WT?+~HB`U9}D|>qf@a({}yhHT>bZ()T#Q
z<9k;)CH$`B>K52hH*p_Ph-a5xP?3p2C9Eo&3P`Sv$Cum^q7s%g_wIF3>Q6Y(B3x}u
z-3PvM=J5~F2Y?Rb>LC(rAn5<11lz*q+0{n*B&mUIrvVMm)e;^z(~*3Y=mSCMC7`Ro
zj`C+-)G|Bl&zESzdfdZ~8u^FJpjS|b%4Kyu(v;zU?nJ|NLg`UH1rToohOL|U0zhDr
z-HtL(F%@VCFXk#ejH4$Qnz@Y%fI&j|T8_pI6@&KwC^iL?2vHBy*eJOw&kqC_)q$%4
zhcJPRI5QM}oVIk?fFM9uMTHSzVkJ5AJ1Q(YDRCw!JeTgBavDV{ppfl}{^*<b6~^-b
zG`Z-;RDGvzf*vRW6nds94N4<gsR~Lg#?rbAmR1)a<gSB&=X7zF_WnT6XsUF3X6eVX
zUm(~sO)(H3YZ3*=o$TU4TyZ5_sw?gZt?{kAXDeKBW;bRs4Fa~){n0+m|J(A2fdral
z64P=w((bgGah#?B{({J4JN59Ei#2fc`s(0MfRPc=F$N}&uxu1nT0W*|3>$E@5h3gQ
z7ZH`L$n)OW5Wg>M2`^*0L%BHZ9}ft8&$7>PkJKuMZRL`Q88-@2Mk2H@nWTFPt5S+F
zLsHzguCWJ@m*dt_$Y?9MPR9%2ZivX+A@%WHx}bu}?mB6wufL%mOccc8b=|TB+Q{GA
zZhnp!1zYQl3})ZmWtz2Umtudiq(x~`fywQMn0h!I_(C6z^7u&~X}5H%u_J;?jT1)h
z)JcCcc(qkZcuVYIsjsrR4uv(klxqC5pVaib0cm-rseQuh8gxy%x4l-Ld@rBX+@0j!
zff+|)Lp1?l9x4YU$o~%vPtYMYop$Ob0Hh25UxJ56k@&OI*n~FT87e{r=<671H{5b~
F_Fv-s=P3XH

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_calendar/lark_calendar.py b/api/core/tools/provider/builtin/lark_calendar/lark_calendar.py
new file mode 100644
index 0000000000..871de69cc1
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/lark_calendar.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkCalendarProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_calendar/lark_calendar.yaml b/api/core/tools/provider/builtin/lark_calendar/lark_calendar.yaml
new file mode 100644
index 0000000000..72c41e36c0
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/lark_calendar.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_calendar
+  label:
+    en_US: Lark Calendar
+    zh_Hans: Lark 日历
+  description:
+    en_US: |
+      Lark calendar, requires the following permissions: calendar:calendar:read、calendar:calendar、contact:user.id:readonly.
+    zh_Hans: |
+      Lark 日历，需要开通以下权限: calendar:calendar:read、calendar:calendar、contact:user.id:readonly。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.py b/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.py
new file mode 100644
index 0000000000..f5929893dd
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class AddEventAttendeesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        event_id = tool_parameters.get("event_id")
+        attendee_phone_or_email = tool_parameters.get("attendee_phone_or_email")
+        need_notification = tool_parameters.get("need_notification", True)
+
+        res = client.add_event_attendees(event_id, attendee_phone_or_email, need_notification)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.yaml
new file mode 100644
index 0000000000..9d7a131907
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/add_event_attendees.yaml
@@ -0,0 +1,54 @@
+identity:
+  name: add_event_attendees
+  author: Doug Lea
+  label:
+    en_US: Add Event Attendees
+    zh_Hans: 添加日程参会人
+description:
+  human:
+    en_US: Add Event Attendees
+    zh_Hans: 添加日程参会人
+  llm: A tool for adding attendees to events in Lark. (在 Lark 中添加日程参会人)
+parameters:
+  - name: event_id
+    type: string
+    required: true
+    label:
+      en_US: Event ID
+      zh_Hans: 日程 ID
+    human_description:
+      en_US: |
+        The ID of the event, which will be returned when the event is created. For example: fb2a6406-26d6-4c8d-a487-6f0246c94d2f_0.
+      zh_Hans: |
+        创建日程时会返回日程 ID。例如: fb2a6406-26d6-4c8d-a487-6f0246c94d2f_0。
+    llm_description: |
+      日程 ID，创建日程时会返回日程 ID。例如: fb2a6406-26d6-4c8d-a487-6f0246c94d2f_0。
+    form: llm
+
+  - name: need_notification
+    type: boolean
+    required: false
+    default: true
+    label:
+      en_US: Need Notification
+      zh_Hans: 是否需要通知
+    human_description:
+      en_US: |
+        Whether to send a Bot notification to attendees. true: send, false: do not send.
+      zh_Hans: |
+        是否给参与人发送 Bot 通知，true: 发送，false: 不发送。
+    llm_description: |
+      是否给参与人发送 Bot 通知，true: 发送，false: 不发送。
+    form: form
+
+  - name: attendee_phone_or_email
+    type: string
+    required: true
+    label:
+      en_US: Attendee Phone or Email
+      zh_Hans: 参会人电话或邮箱
+    human_description:
+      en_US: The list of attendee emails or phone numbers, separated by commas.
+      zh_Hans: 日程参会人邮箱或者手机号列表，使用逗号分隔。
+    llm_description: 日程参会人邮箱或者手机号列表，使用逗号分隔。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/create_event.py b/api/core/tools/provider/builtin/lark_calendar/tools/create_event.py
new file mode 100644
index 0000000000..8a0726008c
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/create_event.py
@@ -0,0 +1,26 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateEventTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        summary = tool_parameters.get("summary")
+        description = tool_parameters.get("description")
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        attendee_ability = tool_parameters.get("attendee_ability")
+        need_notification = tool_parameters.get("need_notification", True)
+        auto_record = tool_parameters.get("auto_record", False)
+
+        res = client.create_event(
+            summary, description, start_time, end_time, attendee_ability, need_notification, auto_record
+        )
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/create_event.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/create_event.yaml
new file mode 100644
index 0000000000..b738736e63
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/create_event.yaml
@@ -0,0 +1,119 @@
+identity:
+  name: create_event
+  author: Doug Lea
+  label:
+    en_US: Create Event
+    zh_Hans: 创建日程
+description:
+  human:
+    en_US: Create Event
+    zh_Hans: 创建日程
+  llm: A tool for creating events in Lark.(创建 Lark 日程)
+parameters:
+  - name: summary
+    type: string
+    required: false
+    label:
+      en_US: Summary
+      zh_Hans: 日程标题
+    human_description:
+      en_US: The title of the event. If not filled, the event title will display (No Subject).
+      zh_Hans: 日程标题，若不填则日程标题显示 (无主题)。
+    llm_description: 日程标题，若不填则日程标题显示 (无主题)。
+    form: llm
+
+  - name: description
+    type: string
+    required: false
+    label:
+      en_US: Description
+      zh_Hans: 日程描述
+    human_description:
+      en_US: The description of the event.
+      zh_Hans: 日程描述。
+    llm_description: 日程描述。
+    form: llm
+
+  - name: need_notification
+    type: boolean
+    required: false
+    default: true
+    label:
+      en_US: Need Notification
+      zh_Hans: 是否发送通知
+    human_description:
+      en_US: |
+        Whether to send a bot message when the event is created, true: send, false: do not send.
+      zh_Hans: 创建日程时是否发送 bot 消息，true：发送，false：不发送。
+    llm_description: 创建日程时是否发送 bot 消息，true：发送，false：不发送。
+    form: form
+
+  - name: start_time
+    type: string
+    required: true
+    label:
+      en_US: Start Time
+      zh_Hans: 开始时间
+    human_description:
+      en_US: |
+        The start time of the event, format: 2006-01-02 15:04:05.
+      zh_Hans: 日程开始时间，格式：2006-01-02 15:04:05。
+    llm_description: 日程开始时间，格式：2006-01-02 15:04:05。
+    form: llm
+
+  - name: end_time
+    type: string
+    required: true
+    label:
+      en_US: End Time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: |
+        The end time of the event, format: 2006-01-02 15:04:05.
+      zh_Hans: 日程结束时间，格式：2006-01-02 15:04:05。
+    llm_description: 日程结束时间，格式：2006-01-02 15:04:05。
+    form: llm
+
+  - name: attendee_ability
+    type: select
+    required: false
+    options:
+      - value: none
+        label:
+          en_US: none
+          zh_Hans: 无
+      - value: can_see_others
+        label:
+          en_US: can_see_others
+          zh_Hans: 可以查看参与人列表
+      - value: can_invite_others
+        label:
+          en_US: can_invite_others
+          zh_Hans: 可以邀请其它参与人
+      - value: can_modify_event
+        label:
+          en_US: can_modify_event
+          zh_Hans: 可以编辑日程
+    default: "none"
+    label:
+      en_US: attendee_ability
+      zh_Hans: 参会人权限
+    human_description:
+      en_US: Attendee ability, optional values are none, can_see_others, can_invite_others, can_modify_event, with a default value of none.
+      zh_Hans: 参会人权限，可选值有无、可以查看参与人列表、可以邀请其它参与人、可以编辑日程，默认值为无。
+    llm_description: 参会人权限，可选值有无、可以查看参与人列表、可以邀请其它参与人、可以编辑日程，默认值为无。
+    form: form
+
+  - name: auto_record
+    type: boolean
+    required: false
+    default: false
+    label:
+      en_US: Auto Record
+      zh_Hans: 自动录制
+    human_description:
+      en_US: |
+        Whether to enable automatic recording, true: enabled, automatically record when the meeting starts; false: not enabled.
+      zh_Hans: 是否开启自动录制，true：开启，会议开始后自动录制；false：不开启。
+    llm_description: 是否开启自动录制，true：开启，会议开始后自动录制；false：不开启。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.py b/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.py
new file mode 100644
index 0000000000..0e4ceac5e5
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class DeleteEventTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        event_id = tool_parameters.get("event_id")
+        need_notification = tool_parameters.get("need_notification", True)
+
+        res = client.delete_event(event_id, need_notification)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.yaml
new file mode 100644
index 0000000000..cdd6d7e1bb
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/delete_event.yaml
@@ -0,0 +1,38 @@
+identity:
+  name: delete_event
+  author: Doug Lea
+  label:
+    en_US: Delete Event
+    zh_Hans: 删除日程
+description:
+  human:
+    en_US: Delete Event
+    zh_Hans: 删除日程
+  llm: A tool for deleting events in Lark.(在 Lark 中删除日程)
+parameters:
+  - name: event_id
+    type: string
+    required: true
+    label:
+      en_US: Event ID
+      zh_Hans: 日程 ID
+    human_description:
+      en_US: |
+        The ID of the event, for example: e8b9791c-39ae-4908-8ad8-66b13159b9fb_0.
+      zh_Hans: 日程 ID，例如：e8b9791c-39ae-4908-8ad8-66b13159b9fb_0。
+    llm_description: 日程 ID，例如：e8b9791c-39ae-4908-8ad8-66b13159b9fb_0。
+    form: llm
+
+  - name: need_notification
+    type: boolean
+    required: false
+    default: true
+    label:
+      en_US: Need Notification
+      zh_Hans: 是否需要通知
+    human_description:
+      en_US: |
+        Indicates whether to send bot notifications to event participants upon deletion. true: send, false: do not send.
+      zh_Hans: 删除日程是否给日程参与人发送 bot 通知，true：发送，false：不发送。
+    llm_description: 删除日程是否给日程参与人发送 bot 通知，true：发送，false：不发送。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.py b/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.py
new file mode 100644
index 0000000000..d315bf35f0
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.py
@@ -0,0 +1,18 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetPrimaryCalendarTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.get_primary_calendar(user_id_type)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.yaml
new file mode 100644
index 0000000000..fe61594770
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/get_primary_calendar.yaml
@@ -0,0 +1,37 @@
+identity:
+  name: get_primary_calendar
+  author: Doug Lea
+  label:
+    en_US: Get Primary Calendar
+    zh_Hans: 查询主日历信息
+description:
+  human:
+    en_US: Get Primary Calendar
+    zh_Hans: 查询主日历信息
+  llm: A tool for querying primary calendar information in Lark.(在 Lark 中查询主日历信息)
+parameters:
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/list_events.py b/api/core/tools/provider/builtin/lark_calendar/tools/list_events.py
new file mode 100644
index 0000000000..d74cc049d3
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/list_events.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ListEventsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        page_token = tool_parameters.get("page_token")
+        page_size = tool_parameters.get("page_size")
+
+        res = client.list_events(start_time, end_time, page_token, page_size)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/list_events.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/list_events.yaml
new file mode 100644
index 0000000000..cef332f527
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/list_events.yaml
@@ -0,0 +1,62 @@
+identity:
+  name: list_events
+  author: Doug Lea
+  label:
+    en_US: List Events
+    zh_Hans: 获取日程列表
+description:
+  human:
+    en_US: List Events
+    zh_Hans: 获取日程列表
+  llm: A tool for listing events in Lark.(在 Lark 中获取日程列表)
+parameters:
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 开始时间
+    human_description:
+      en_US: |
+        The start time, defaults to 0:00 of the current day if not provided, format: 2006-01-02 15:04:05.
+      zh_Hans: 开始时间，不传值时默认当天 0 点时间，格式为:2006-01-02 15:04:05。
+    llm_description: 开始时间，不传值时默认当天 0 点时间，格式为:2006-01-02 15:04:05。
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: |
+        The end time, defaults to 23:59 of the current day if not provided, format: 2006-01-02 15:04:05.
+      zh_Hans: 结束时间，不传值时默认当天 23:59 分时间，格式为:2006-01-02 15:04:05。
+    llm_description: 结束时间，不传值时默认当天 23:59 分时间，格式为:2006-01-02 15:04:05。
+    form: llm
+
+  - name: page_size
+    type: number
+    required: false
+    default: 50
+    label:
+      en_US: Page Size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 50, and the value range is [50,1000].
+      zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 50，取值范围为 [50,1000]。
+    llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 50，取值范围为 [50,1000]。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: Page Token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: The pagination token. Leave it blank for the first request, indicating to start traversing from the beginning; when the pagination query result has more items, a new page_token will be returned simultaneously, which can be used to obtain the query result in the next traversal.
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/search_events.py b/api/core/tools/provider/builtin/lark_calendar/tools/search_events.py
new file mode 100644
index 0000000000..a20038e47d
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/search_events.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class SearchEventsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        query = tool_parameters.get("query")
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        page_token = tool_parameters.get("page_token")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+        page_size = tool_parameters.get("page_size", 20)
+
+        res = client.search_events(query, start_time, end_time, page_token, user_id_type, page_size)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/search_events.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/search_events.yaml
new file mode 100644
index 0000000000..4d4f8819c1
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/search_events.yaml
@@ -0,0 +1,100 @@
+identity:
+  name: search_events
+  author: Doug Lea
+  label:
+    en_US: Search Events
+    zh_Hans: 搜索日程
+description:
+  human:
+    en_US: Search Events
+    zh_Hans: 搜索日程
+  llm: A tool for searching events in Lark.(在 Lark 中搜索日程)
+parameters:
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: query
+    type: string
+    required: true
+    label:
+      en_US: Query
+      zh_Hans: 搜索关键字
+    human_description:
+      en_US: The search keyword used for fuzzy searching event names, with a maximum input of 200 characters.
+      zh_Hans: 用于模糊查询日程名称的搜索关键字，最大输入 200 字符。
+    llm_description: 用于模糊查询日程名称的搜索关键字，最大输入 200 字符。
+    form: llm
+
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 开始时间
+    human_description:
+      en_US: |
+        The start time, defaults to 0:00 of the current day if not provided, format: 2006-01-02 15:04:05.
+      zh_Hans: 开始时间，不传值时默认当天 0 点时间，格式为:2006-01-02 15:04:05。
+    llm_description: 开始时间，不传值时默认当天 0 点时间，格式为:2006-01-02 15:04:05。
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: |
+        The end time, defaults to 23:59 of the current day if not provided, format: 2006-01-02 15:04:05.
+      zh_Hans: 结束时间，不传值时默认当天 23:59 分时间，格式为:2006-01-02 15:04:05。
+    llm_description: 结束时间，不传值时默认当天 23:59 分时间，格式为:2006-01-02 15:04:05。
+    form: llm
+
+  - name: page_size
+    type: number
+    required: false
+    default: 20
+    label:
+      en_US: Page Size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [10,100].
+      zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [10,100]。
+    llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [10,100]。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: Page Token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: The pagination token. Leave it blank for the first request, indicating to start traversing from the beginning; when the pagination query result has more items, a new page_token will be returned simultaneously, which can be used to obtain the query result in the next traversal.
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/update_event.py b/api/core/tools/provider/builtin/lark_calendar/tools/update_event.py
new file mode 100644
index 0000000000..a04029377f
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/update_event.py
@@ -0,0 +1,24 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class UpdateEventTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        event_id = tool_parameters.get("event_id")
+        summary = tool_parameters.get("summary")
+        description = tool_parameters.get("description")
+        need_notification = tool_parameters.get("need_notification", True)
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        auto_record = tool_parameters.get("auto_record", False)
+
+        res = client.update_event(event_id, summary, description, need_notification, start_time, end_time, auto_record)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_calendar/tools/update_event.yaml b/api/core/tools/provider/builtin/lark_calendar/tools/update_event.yaml
new file mode 100644
index 0000000000..b9992e5b03
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_calendar/tools/update_event.yaml
@@ -0,0 +1,100 @@
+identity:
+  name: update_event
+  author: Doug Lea
+  label:
+    en_US: Update Event
+    zh_Hans: 更新日程
+description:
+  human:
+    en_US: Update Event
+    zh_Hans: 更新日程
+  llm: A tool for updating events in Lark.(更新 Lark 中的日程)
+parameters:
+  - name: event_id
+    type: string
+    required: true
+    label:
+      en_US: Event ID
+      zh_Hans: 日程 ID
+    human_description:
+      en_US: |
+        The ID of the event, for example: e8b9791c-39ae-4908-8ad8-66b13159b9fb_0.
+      zh_Hans: 日程 ID，例如：e8b9791c-39ae-4908-8ad8-66b13159b9fb_0。
+    llm_description: 日程 ID，例如：e8b9791c-39ae-4908-8ad8-66b13159b9fb_0。
+    form: llm
+
+  - name: summary
+    type: string
+    required: false
+    label:
+      en_US: Summary
+      zh_Hans: 日程标题
+    human_description:
+      en_US: The title of the event.
+      zh_Hans: 日程标题。
+    llm_description: 日程标题。
+    form: llm
+
+  - name: description
+    type: string
+    required: false
+    label:
+      en_US: Description
+      zh_Hans: 日程描述
+    human_description:
+      en_US: The description of the event.
+      zh_Hans: 日程描述。
+    llm_description: 日程描述。
+    form: llm
+
+  - name: need_notification
+    type: boolean
+    required: false
+    label:
+      en_US: Need Notification
+      zh_Hans: 是否发送通知
+    human_description:
+      en_US: |
+        Whether to send a bot message when the event is updated, true: send, false: do not send.
+      zh_Hans: 更新日程时是否发送 bot 消息，true：发送，false：不发送。
+    llm_description: 更新日程时是否发送 bot 消息，true：发送，false：不发送。
+    form: form
+
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 开始时间
+    human_description:
+      en_US: |
+        The start time of the event, format: 2006-01-02 15:04:05.
+      zh_Hans: 日程开始时间，格式：2006-01-02 15:04:05。
+    llm_description: 日程开始时间，格式：2006-01-02 15:04:05。
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: |
+        The end time of the event, format: 2006-01-02 15:04:05.
+      zh_Hans: 日程结束时间，格式：2006-01-02 15:04:05。
+    llm_description: 日程结束时间，格式：2006-01-02 15:04:05。
+    form: llm
+
+  - name: auto_record
+    type: boolean
+    required: false
+    label:
+      en_US: Auto Record
+      zh_Hans: 自动录制
+    human_description:
+      en_US: |
+        Whether to enable automatic recording, true: enabled, automatically record when the meeting starts; false: not enabled.
+      zh_Hans: 是否开启自动录制，true：开启，会议开始后自动录制；false：不开启。
+    llm_description: 是否开启自动录制，true：开启，会议开始后自动录制；false：不开启。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_document/_assets/icon.svg b/api/core/tools/provider/builtin/lark_document/_assets/icon.svg
new file mode 100644
index 0000000000..5a0a6416b3
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/_assets/icon.svg
@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" version="1.1" width="64px" height="64px" style="shape-rendering:geometricPrecision; text-rendering:geometricPrecision; image-rendering:optimizeQuality; fill-rule:evenodd; clip-rule:evenodd" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g><path style="opacity:1" fill="#fefefe" d="M -0.5,-0.5 C 20.8333,-0.5 42.1667,-0.5 63.5,-0.5C 63.5,20.8333 63.5,42.1667 63.5,63.5C 42.1667,63.5 20.8333,63.5 -0.5,63.5C -0.5,42.1667 -0.5,20.8333 -0.5,-0.5 Z"/></g>
+<g><path style="opacity:1" fill="#346df3" d="M 47.5,33.5 C 43.3272,29.8779 38.9939,29.7112 34.5,33C 32.682,35.4897 30.3487,37.3231 27.5,38.5C 23.5003,43.5136 24.167,47.847 29.5,51.5C 24.1563,51.666 18.8229,51.4994 13.5,51C 13,50.5 12.5,50 12,49.5C 11.3333,36.8333 11.3333,24.1667 12,11.5C 12.5,11 13,10.5 13.5,10C 24.1667,9.33333 34.8333,9.33333 45.5,10C 46,10.5 46.5,11 47,11.5C 47.4997,18.8258 47.6663,26.1591 47.5,33.5 Z"/></g>
+<g><path style="opacity:1" fill="#f9fafe" d="M 20.5,19.5 C 25.1785,19.3342 29.8452,19.5008 34.5,20C 35.8333,21 35.8333,22 34.5,23C 29.8333,23.6667 25.1667,23.6667 20.5,23C 19.3157,21.8545 19.3157,20.6879 20.5,19.5 Z"/></g>
+<g><path style="opacity:1" fill="#f3f6fe" d="M 20.5,27.5 C 22.5273,27.3379 24.5273,27.5045 26.5,28C 27.8333,29 27.8333,30 26.5,31C 24.5,31.6667 22.5,31.6667 20.5,31C 19.3157,29.8545 19.3157,28.6879 20.5,27.5 Z"/></g>
+<g><path style="opacity:1" fill="#36d4c1" d="M 47.5,33.5 C 48.7298,35.2972 49.3964,37.2972 49.5,39.5C 51.3904,39.2965 52.8904,39.9632 54,41.5C 55.1825,45.2739 54.3492,48.4406 51.5,51C 44.1742,51.4997 36.8409,51.6663 29.5,51.5C 24.167,47.847 23.5003,43.5136 27.5,38.5C 30.3487,37.3231 32.682,35.4897 34.5,33C 38.9939,29.7112 43.3272,29.8779 47.5,33.5 Z"/></g>
+</svg>
diff --git a/api/core/tools/provider/builtin/lark_document/lark_document.py b/api/core/tools/provider/builtin/lark_document/lark_document.py
new file mode 100644
index 0000000000..b128327602
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/lark_document.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkDocumentProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_document/lark_document.yaml b/api/core/tools/provider/builtin/lark_document/lark_document.yaml
new file mode 100644
index 0000000000..0cb4ae1d62
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/lark_document.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_document
+  label:
+    en_US: Lark Cloud Document
+    zh_Hans: Lark 云文档
+  description:
+    en_US: |
+      Lark cloud document, requires the following permissions: docx:document、drive:drive、docs:document.content:read.
+    zh_Hans: |
+      Lark 云文档，需要开通以下权限: docx:document、drive:drive、docs:document.content:read。
+  icon: icon.svg
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_document/tools/create_document.py b/api/core/tools/provider/builtin/lark_document/tools/create_document.py
new file mode 100644
index 0000000000..2b1dae0db5
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/create_document.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateDocumentTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        title = tool_parameters.get("title")
+        content = tool_parameters.get("content")
+        folder_token = tool_parameters.get("folder_token")
+
+        res = client.create_document(title, content, folder_token)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_document/tools/create_document.yaml b/api/core/tools/provider/builtin/lark_document/tools/create_document.yaml
new file mode 100644
index 0000000000..37a1e23041
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/create_document.yaml
@@ -0,0 +1,48 @@
+identity:
+  name: create_document
+  author: Doug Lea
+  label:
+    en_US: Create Lark document
+    zh_Hans: 创建 Lark 文档
+description:
+  human:
+    en_US: Create Lark document
+    zh_Hans: 创建 Lark 文档，支持创建空文档和带内容的文档，支持 markdown 语法创建。应用需要开启机器人能力(https://open.larksuite.com/document/faq/trouble-shooting/how-to-enable-bot-ability)。
+  llm: A tool for creating Lark documents.
+parameters:
+  - name: title
+    type: string
+    required: false
+    label:
+      en_US: Document title
+      zh_Hans: 文档标题
+    human_description:
+      en_US: Document title, only supports plain text content.
+      zh_Hans: 文档标题，只支持纯文本内容。
+    llm_description: 文档标题，只支持纯文本内容，可以为空。
+    form: llm
+
+  - name: content
+    type: string
+    required: false
+    label:
+      en_US: Document content
+      zh_Hans: 文档内容
+    human_description:
+      en_US: Document content, supports markdown syntax, can be empty.
+      zh_Hans: 文档内容，支持 markdown 语法，可以为空。
+    llm_description: 文档内容，支持 markdown 语法，可以为空。
+    form: llm
+
+  - name: folder_token
+    type: string
+    required: false
+    label:
+      en_US: folder_token
+      zh_Hans: 文档所在文件夹的 Token
+    human_description:
+      en_US: |
+        The token of the folder where the document is located. If it is not passed or is empty, it means the root directory. For Example: https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd
+      zh_Hans: 文档所在文件夹的 Token，不传或传空表示根目录。例如：https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd。
+    llm_description: 文档所在文件夹的 Token，不传或传空表示根目录。例如：https://lark-japan.jp.larksuite.com/drive/folder/Lf8uf6BoAlWkUfdGtpMjUV0PpZd。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_document/tools/get_document_content.py b/api/core/tools/provider/builtin/lark_document/tools/get_document_content.py
new file mode 100644
index 0000000000..d15211b57e
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/get_document_content.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetDocumentRawContentTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        document_id = tool_parameters.get("document_id")
+        mode = tool_parameters.get("mode", "markdown")
+        lang = tool_parameters.get("lang", "0")
+
+        res = client.get_document_content(document_id, mode, lang)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_document/tools/get_document_content.yaml b/api/core/tools/provider/builtin/lark_document/tools/get_document_content.yaml
new file mode 100644
index 0000000000..fd6a033bfd
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/get_document_content.yaml
@@ -0,0 +1,70 @@
+identity:
+  name: get_document_content
+  author: Doug Lea
+  label:
+    en_US: Get Lark Cloud Document Content
+    zh_Hans: 获取 Lark 云文档的内容
+description:
+  human:
+    en_US: Get lark cloud document content
+    zh_Hans: 获取 Lark 云文档的内容
+  llm: A tool for retrieving content from Lark cloud documents.
+parameters:
+  - name: document_id
+    type: string
+    required: true
+    label:
+      en_US: document_id
+      zh_Hans: Lark 文档的唯一标识
+    human_description:
+      en_US: Unique identifier for a Lark document. You can also input the document's URL.
+      zh_Hans: Lark 文档的唯一标识，支持输入文档的 URL。
+    llm_description: Lark 文档的唯一标识，支持输入文档的 URL。
+    form: llm
+
+  - name: mode
+    type: select
+    required: false
+    options:
+      - value: text
+        label:
+          en_US: text
+          zh_Hans: text
+      - value: markdown
+        label:
+          en_US: markdown
+          zh_Hans: markdown
+    default: "markdown"
+    label:
+      en_US: mode
+      zh_Hans: 文档返回格式
+    human_description:
+      en_US: Format of the document return, optional values are text, markdown, can be empty, default is markdown.
+      zh_Hans: 文档返回格式，可选值有 text、markdown，可以为空，默认值为 markdown。
+    llm_description: 文档返回格式，可选值有 text、markdown，可以为空，默认值为 markdown。
+    form: form
+
+  - name: lang
+    type: select
+    required: false
+    options:
+      - value: "0"
+        label:
+          en_US: User's default name
+          zh_Hans: 用户的默认名称
+      - value: "1"
+        label:
+          en_US: User's English name
+          zh_Hans: 用户的英文名称
+    default: "0"
+    label:
+      en_US: lang
+      zh_Hans: 指定@用户的语言
+    human_description:
+      en_US: |
+        Specifies the language for MentionUser, optional values are [0, 1]. 0: User's default name, 1: User's English name, default is 0.
+      zh_Hans: |
+        指定返回的 MentionUser，即@用户的语言，可选值有 [0,1]。0: 该用户的默认名称，1: 该用户的英文名称，默认值为 0。
+    llm_description: |
+      指定返回的 MentionUser，即@用户的语言，可选值有 [0,1]。0: 该用户的默认名称，1: 该用户的英文名称，默认值为 0。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.py b/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.py
new file mode 100644
index 0000000000..b96a87489e
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ListDocumentBlockTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        document_id = tool_parameters.get("document_id")
+        page_token = tool_parameters.get("page_token", "")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+        page_size = tool_parameters.get("page_size", 500)
+
+        res = client.list_document_blocks(document_id, page_token, user_id_type, page_size)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.yaml b/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.yaml
new file mode 100644
index 0000000000..08b673e0ae
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/list_document_blocks.yaml
@@ -0,0 +1,74 @@
+identity:
+  name: list_document_blocks
+  author: Doug Lea
+  label:
+    en_US: List Lark Document Blocks
+    zh_Hans: 获取 Lark 文档所有块
+description:
+  human:
+    en_US: List lark document blocks
+    zh_Hans: 获取 Lark 文档所有块的富文本内容并分页返回
+  llm: A tool to get all blocks of Lark documents
+parameters:
+  - name: document_id
+    type: string
+    required: true
+    label:
+      en_US: document_id
+      zh_Hans: Lark 文档的唯一标识
+    human_description:
+      en_US: Unique identifier for a Lark document. You can also input the document's URL.
+      zh_Hans: Lark 文档的唯一标识，支持输入文档的 URL。
+    llm_description: Lark 文档的唯一标识，支持输入文档的 URL。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: page_size
+    type: number
+    required: false
+    default: 500
+    label:
+      en_US: page_size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: Paging size, the default and maximum value is 500.
+      zh_Hans: 分页大小, 默认值和最大值为 500。
+    llm_description: 分页大小, 表示一次请求最多返回多少条数据，默认值和最大值为 500。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: page_token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: Pagination token used to navigate through query results, allowing retrieval of additional items in subsequent requests.
+      zh_Hans: 分页标记，用于分页查询结果，以便下次遍历时获取更多项。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_document/tools/write_document.py b/api/core/tools/provider/builtin/lark_document/tools/write_document.py
new file mode 100644
index 0000000000..888e0e39fc
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/write_document.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateDocumentTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        document_id = tool_parameters.get("document_id")
+        content = tool_parameters.get("content")
+        position = tool_parameters.get("position", "end")
+
+        res = client.write_document(document_id, content, position)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_document/tools/write_document.yaml b/api/core/tools/provider/builtin/lark_document/tools/write_document.yaml
new file mode 100644
index 0000000000..9cdf034ed0
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_document/tools/write_document.yaml
@@ -0,0 +1,57 @@
+identity:
+  name: write_document
+  author: Doug Lea
+  label:
+    en_US: Write Document
+    zh_Hans: 在 Lark 文档中新增内容
+description:
+  human:
+    en_US: Adding new content to Lark documents
+    zh_Hans: 在 Lark 文档中新增内容
+  llm: A tool for adding new content to Lark documents.
+parameters:
+  - name: document_id
+    type: string
+    required: true
+    label:
+      en_US: document_id
+      zh_Hans: Lark 文档的唯一标识
+    human_description:
+      en_US: Unique identifier for a Lark document. You can also input the document's URL.
+      zh_Hans: Lark 文档的唯一标识，支持输入文档的 URL。
+    llm_description: Lark 文档的唯一标识，支持输入文档的 URL。
+    form: llm
+
+  - name: content
+    type: string
+    required: true
+    label:
+      en_US: Plain text or Markdown content
+      zh_Hans: 纯文本或 Markdown 内容
+    human_description:
+      en_US: Plain text or Markdown content. Note that embedded tables in the document should not have merged cells.
+      zh_Hans: 纯文本或 Markdown 内容。注意文档的内嵌套表格不允许有单元格合并。
+    llm_description: 纯文本或 Markdown 内容，注意文档的内嵌套表格不允许有单元格合并。
+    form: llm
+
+  - name: position
+    type: select
+    required: false
+    options:
+      - value: start
+        label:
+          en_US: document start
+          zh_Hans: 文档开始
+      - value: end
+        label:
+          en_US: document end
+          zh_Hans: 文档结束
+    default: "end"
+    label:
+      en_US: position
+      zh_Hans: 内容添加位置
+    human_description:
+      en_US: Content insertion position, optional values are start, end. 'start' means adding content at the beginning of the document; 'end' means adding content at the end of the document. The default value is end.
+      zh_Hans: 内容添加位置，可选值有 start、end。start 表示在文档开头添加内容；end 表示在文档结尾添加内容，默认值为 end。
+    llm_description: 内容添加位置，可选值有 start、end。start 表示在文档开头添加内容；end 表示在文档结尾添加内容，默认值为 end。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/_assets/icon.png b/api/core/tools/provider/builtin/lark_message_and_group/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..0dfd58a9d512fd0f0481e32f913ba978d6219002
GIT binary patch
literal 4482
zcmV-|5q<87P)<h;3K|Lk000e1NJLTq002M$002M;1^@s6s%dfF000q5Nkl<ZcwX&U
z`FB-Cwyt83p#c$rZnTjoAc%sDB8UjcBx9IELc(PZQwXCF`WduEy0u%d)klj4-e1u6
zFAYu@Q9%R&QHD(9W+Y*}M1Mc~eS25cIp<zb;61v<wWzhewRi2>Rdv3-chxyJQR+^+
z)9$o8?M}OOEjy}OuBuXzHD}e<HRtp{R@La}>RJ_BU88ArEkn`j8Wjcd*=O~6H9NaT
zB_*!&t7<~JM&+&qsa0nct*p`BN-L!=x)K`#BD%9130t)~2K%oo7)(41#8;eEm$R!C
zF;-Z4O`H{Y+0~j>h=l)H?KxYv=$h@jZu>xFRjXVemQ|(bDpV`aNNpsryh?e?VP65F
zfv*nvgzf%1xrzJ!;w>|(lCmn*v1OHtmVs0twb6m)m0F~Q_3_7jziVN6ND8WSH1l?;
zhNV?%_L2$}TT-rQX@w@}fc97}^<nIOzcR2JU=D)i#^GyqndjTh82ibb7%(yMdH$m<
zvr^?Q260AanUMn4@u051q+ADMak}dpTV{DMNBSUt^_=+mndS9<nIn-|u5uSe4e><|
zAmHm|+V*{$m9`%Z&2uoeU=8luJu-Q{V69*e$c6vuAS^6V$qVsd3!>UH5-7!kM78K_
zb9P^saWf}uw|<_?Ei+@*$8)~%YqCuT{W9|m3(M8)#OM6CG`{fWm#P>*S3}dhQsvD%
zWBM$)FM6pKIUfXA7e5O(U*_?+%oCqr9oTs+yr1)%Z`O|Svn4+7EpKzr=;&N*91raU
z(h!(iVsS>XE$nZ74d&S~C+C=PS;zM?>x6s*+nkUe^ka$7drMoSSmn;a#u0(`0G6g-
zEOu{>>3bI9C$=-}Z~V*}GT)39Yhr1@<7{(l${wCuKeU&FtmE6y{F}ZLA}Fuf#oA+J
z6e(iN#>N?JW8K<k7aOoWqgVywc1}i-A!B@$XZ*G3rXO0L$J`!Q%`s~VO58ZFqc2B}
zsg{Y)|Nn03MQRs7fR_#u>4l<`k(zB*w|&+NwLb>pG>hj3{!W)M*PrFjb!*4(X?)!}
z#>d*Mjx~yJaTP>PtE5>VlUAs_v;swfLhb9b7(5=b&njeHc_PKmx(@sj*Wj^&&>9Zb
z7M^GHz}Mfy?OSw<t048Xilk!knFZRLd0J7bLxF+VSP$9RX6zdCl*e|S?<Cve)(Wrb
z&x60f*xWp^h4u-roqAe3+pVjD8K=~-lv9dQusGm3%wT!S+GJcA8zewtb8tR!Hf86z
zwT#`Y5pQ#t$<M$(r{eu*_<L%M$=dmeJpNBD<(NvEep1o&lbWVuQ===R1H75(t3YR8
z=Xa7H$n?#4Fc#SD+E^nP53doxpW}SoGs#@*W9`Ye;uuW(PDQ4kP&Cbgb=RMIQhQTR
zXwTR+P2>Ja*LQQYKgZZi-(T;f^4R8d1GA1>%imXg!*h9m+dpa7KgsJG`}7mH;OlVG
zakXvA_ln|x4JMyZUZU>X5-=G)Q@+<h|9kE`o#$v-C+uUO&0MT&)-bj>9|Ui*+v9{8
zll9GWZB68J({0lCDj)DBeWxf6Y&-7y+VolaK32!W2;#=%A)OwqEq-=w*eA=n)&?Yh
z7p&vYhc5>paebbbd|drAk<U%HiN{qm`MBoI2@Cp|hzFeb9dBan@Z$|d8oTuqXwQu^
z31f$i*9fg?#=`q(@eA9W|26Dhu(m%xn#kv-+k|fwg)!d1V+3{`=fnMp&UP%`J{}Kl
z=9+Q4maqH!z;6O0Ze2r;&v{BgJ$y7Zz^8Cy{4i>cpUtgkakMQhE^S8(OWILJK`R8b
z!R16gH{Hg4iw6dYaRP0n@{F!M*3G!l+286k-uegg<8|j3jA=7|6p(eS{8lQ6Z+8At
zq>Pe!`21}E+=U-NO%Z?=2*ACxw5%PisCbapRCJ}Bs$R6|>_FOlZrH9ZHKUUfdEIau
zb3_qi>`_f)j<QsN6uapgoADU~Vqrh4i_PfTlXVl;b7Rh*bq@425Kk`-qdtPr5I=kx
z+W@qr%xD_~;C{+3e~8vpK0-Owy(Ivz)DEZD&X1$57bb7r8cV+U*Wz<W)MbDeqmL+W
zw9~&ac<zXXO?!}jpLI7!bYu7D1!KXwX56l0<yd4sm48ow^ejO=K{~Q;0?-6MgPS7&
z_n-pW(6X}jlwHx0)>b}>0Q5ls1|a|=5P<Qt?ZQ-g>r!fTYb=tK$nj^}s3R&m>ae0w
z-zX0tqZ_i)_iZ65(4O(rfgQv~VjZzLpJ0wz-`d8a{F%HEmmgA0zCO@^W+4FSC5;e(
zrU*a_1fUfv;CBeX0|-DT1mH2+Sly3aJUbWx7>NK(KmevA0BN-SQY5+~wxDMs&!23=
z53Bb_98xsmu<`&BJ&eQJ6F?uyIINwo_>XjRxNqmMkNAmh#;g<@DuBQ3ZPqRMB1^v5
zkk7#77b$6m0NhOrN?YRsybl3LLI577byYnOfd2Gi%@EphZZ!Sne6j@K?Mvwhz+Bo9
zTNK?M%S=k-dd)WcklOm}A>|D}qzO7X`RvyghqO2RYqpzyoF2AWUtY_fW7lwVh9i(1
zL?(}4bOI)yCfR554J|kTO%Z_RxC2@u0BvY-+3#st`Gd3)=YL&QPx`O31L&pNVf3f-
zW9Y9JCes@iQ|Rr>8T9wqJbE{lNqJY6Mt8-suYV^D`$|QI9aJRvT6?}8>JMf4wU+)+
z88f=Ev0vPrkRGy|KEKO-@=QEc0#KjQLgdqno8S%53;}3?3TTA@{1yRdPs__Y(kh()
z^*H}8;Qap)7vP_90si&kRC?o5DgqEe0Or%XR~A!V-E!LX<BIElDGte3Er%S`-jJ^}
z4f#q5hFa-^gK#k3HuRwLXQ`#{Y$6XTFYq;Ma24>mPX&1+IRN#620%kfExrqZXoMg%
zK?RuX&pXqa7Dn6AlClnzh4Nor*#iOSN4YhF>E&~y5P%8v`o-z=Cd&W6FVCTOVha&~
zr3gS4?W|i#sDh+K&R4Z(4ygQR@SuagQr_SL=AoZCpe<O(7z5*epLGyt@R+k1KkYeR
z)}7?+&t&3-^#n5unS4WwMj+o9Xo3o8Mj6E|5P()Rue2@R0Qch+_z=#2H(HDLe@^w2
zw5et&z5&M2YZoTbR+RsnmojKO%Kx1!nFzqLAOLl%w<Yqustww&a-ZI>=;{5ML|Vcu
zjXl8N1J=j-U1ba7fcD(F90VpmlZh7?@-quf=F^Jqk^rO@H9_`&g8($g5on13w8k5t
z9o_&PaQ-{f%8JKu{`=A&Q2sCB^|$5RIQk1-f7>pm(pxD1?Xd;4<H`~QV7Vo~igx|5
zCU)JsfRDjbU#b{DKZ*U?8}y}yRG#SjS=XNH2fo(M5<qapKJpv>yye@!9%Ow=0r{DQ
z4FM+K7!`r+7yU+%UVJxRf%oG4|CSb(wnzDQq^$C;_yFpK@_!QNe+a!?I~wJmOt0hi
z{~OBx?aOm<{ufF4zZYa50a(k3Byzs04cw<T4fsNP1HJ^lP&5z-%Q*c@<qiBohsQjl
zYmo+KoF)5@|JHopJMt+)z7Zq>P}l@$3i8cS0WEO;??C|Cp#1Mg`6uD^*9GO@9og@P
z@*jlqA5O0z`>$P?g7QyA`A2a6=cD`=OZM~XvT5fJl6@P1=XWLYy{d5)^#5GPp8Q-<
zJhaIEASXJ=`eXYNeM4Y>_LX_^^Hs>tIBm$M6mS3<TX5!^3HkJrRw)10xcu%z_8-9e
zzZ0_mC@#O=$o@dQ{)XW8AC1>vGG2c(aQg?@-@)0xk|hBMkw+ywe`Ehpd19Z+ePW*`
z1I6$}ehvxrXZz>c>nC>W^O+srr;RVr-%20wnRW`~3mO0>^M-g+A)i{<905S)iw*fX
z__enH<TIlkk^M(-_IuL0$|vygGZ@()L7QvG)2rvFBKv7J`wRT+zX$SpIQu&#0P7&v
zaS&pOTz|6l`%E9}iwEup^v!3fMVJ0(%IgdA{<sg#HqXs*e!kB>{-dTLpHhIbFR<hb
zka;Fw*a9C!OumhfpMR!3Eh>EoXCI|s-UDa9A8n|5igM2mNA|}c`;(CU6l8xkvOgb}
zU#89e3Xork^Ur7BkT(HXPl-H#vh~bYNqzEFtarZh`s8bZPOSI-)JX38wtVgN0eK#C
zb~8tlj9EYSNhB^=2l2F`mbmoprP;-8X>Q4VKKZ55uCxN=S6B9>^;IbS>Y@1b9gUlR
zBEJ2nA^Wp%^Up>07b5%S=6B?GqVx^<=Phm`0Ni@+RgqqM@$jE&>IHHEwCD7mzRuF>
zd&9mmX2E~j=8KMrH!~S81aV|Nqqq&o{|=eIALKiLd`BUlRo<OeRp8ts^MA19H`h*(
z(trIz8gBjwvcCXle=)MZ3;|ey@?R;XAI>~u=MNih{0O+#^JBFMfuI1I9^b1y*MIzz
zz~&IxEk0$N7JbSIF88}Z{2ma$mokc4OV;Ng>vKzfPYcdGNSURbLH-fSD(gY3%KOms
zl>>0@hoJOF%DMm3xhW|9nfUIXjqJ~p(*JvGsZTzSvtP&A4>F&}0jS$>W0zpK_1LR&
zdw@*$j}`U!L=)>_bpJ$qas7|krmW5R2XfreW<okr)RrR1_*~9-NqZ1alB_R|cA;g_
zZaDLg)9Q-8AV1J2|8ng(ke`fSeJS|p$w2nG^cPw3SpoUD(wEE&5QOW$2E*;Kk5w)}
zfcMx(A}zFMAo?|Ib`DFK7ubg!XhL&~@1uFh_WY6uXdyDb=*&a3xb$Hmo>kVJR+jgs
zwG~gG><3BNZ#p{)<T>+`@y<`dxlhNfZ_fSp*b<cfas<HK`gwJ(^h24K0OYW|@&6xm
z1Nlck($ozE9t9x1ZruM+S(q`_dEEKJ$75gH_ry1lsDk?5yy6a&DWtj7yU=n+ysRfm
zzAvq-7(h8yPvgRSHbj03K6huzrT4c>b8zmN{9;^u=G^<_{n9t&^L+9-{DbJ{KZRZE
z`k~70`hlWHK4fTt^gm$T*iGM#TOIQ}wu}5wdtygG>^e@;grjX~S!q{VUe=9PfcUEN
zUbGs-*H-kW^_7Ec$&a9yYR1Tg_vdp{E%|gyenA|0SNeHEp0m$c4`)6HfylYB@=tnS
zwd}G-Q5QULmpz(9XM7;=9{xZB%Jc0ZA12RZEd4og5^F;5`N>@<bDVK0^?sK4Q^@*Y
zdI4E~5n2CZ%~*P+b|Sv>r{cm(_0PP^Jm2^`63)E|K=|AXambCBvpy-2>opq(p!04`
zop&q2!>sQC={>f;hsTVq^B%<@mST4{nP>b3;)gNJbL&hw)&ZmkN~v$GdIse_4COu&
zS$_`1Cz$hoemcmf33)#AiDhrd^O=vo^i0`@vR{|8DUs*Twod<2xgB?_%bj*>uj9WI
zbxHun;_Mya2VdXLQpVWk{Gda|Kt9@Ow|c+FM~$|<P(3u7TRq%n{5jfOGoH58Cey3u
zrs1Zaf%p6@dDXp%SKU?Q<<jG_zt*K^GSApmms6U^@3v{VHKUU@*NlyB;fxD$oONXV
zb(Fk2?{DF}n^&E??U}qi^KoS#U;143d3772H(dI^RD1P&a?e-KO^F)f=9ZVNOUd(P
z#~(RNKJmh{H-6nZ1R`AertIa`|8c|h{}y;Vw5>HMN!u=_Mc<IKj$8iHeEW{)l3yyB
z=N~wJ=0TjZf9=c%<uCe=H+lYlskSv1Y5C^GIok|z&bs@y&$5~4_q_YW4SwLnf9P!V
zr3-@G12_8qza1L?BjcUe(*JSf<#+uR{H~Xh4=z0W<0hf(1L97H^nO(VaE%{2FP0s7
zFP6OvKW&eK{AHYX59BR20PDRxkvTrzL*KX6ZM^($-G=PMwePe$?M}PX?liRj1%y^h
Uj-90R!vFvP07*qoM6N<$g44np;{X5v

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.py b/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.py
new file mode 100644
index 0000000000..de6997b0bf
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkMessageAndGroupProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.yaml b/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.yaml
new file mode 100644
index 0000000000..ad3fe0f361
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/lark_message_and_group.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_message_and_group
+  label:
+    en_US: Lark Message And Group
+    zh_Hans: Lark 消息和群组
+  description:
+    en_US: |
+      Lark message and group, requires the following permissions: im:message、im:message.group_msg.
+    zh_Hans: |
+      Lark 消息和群组，需要开通以下权限: im:message、im:message.group_msg。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.py b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.py
new file mode 100644
index 0000000000..118bac7ab7
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetChatMessagesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        container_id = tool_parameters.get("container_id")
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        page_token = tool_parameters.get("page_token")
+        sort_type = tool_parameters.get("sort_type", "ByCreateTimeAsc")
+        page_size = tool_parameters.get("page_size", 20)
+
+        res = client.get_chat_messages(container_id, start_time, end_time, page_token, sort_type, page_size)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.yaml b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.yaml
new file mode 100644
index 0000000000..965b45a5fb
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_chat_messages.yaml
@@ -0,0 +1,96 @@
+identity:
+  name: get_chat_messages
+  author: Doug Lea
+  label:
+    en_US: Get Chat Messages
+    zh_Hans: 获取指定单聊、群聊的消息历史
+description:
+  human:
+    en_US: Get Chat Messages
+    zh_Hans: 获取指定单聊、群聊的消息历史
+  llm: A tool for getting chat messages from specific one-on-one chats or group chats.(获取指定单聊、群聊的消息历史)
+parameters:
+  - name: container_id
+    type: string
+    required: true
+    label:
+      en_US: Container Id
+      zh_Hans: 群聊或单聊的 ID
+    human_description:
+      en_US: The ID of the group chat or single chat. Refer to the group ID description for how to obtain it. https://open.larkoffice.com/document/server-docs/group/chat/chat-id-description
+      zh_Hans: 群聊或单聊的 ID，获取方式参见群 ID 说明。https://open.larkoffice.com/document/server-docs/group/chat/chat-id-description
+    llm_description: 群聊或单聊的 ID，获取方式参见群 ID 说明。https://open.larkoffice.com/document/server-docs/group/chat/chat-id-description
+    form: llm
+
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 起始时间
+    human_description:
+      en_US: The start time for querying historical messages, formatted as "2006-01-02 15:04:05".
+      zh_Hans: 待查询历史信息的起始时间，格式为 "2006-01-02 15:04:05"。
+    llm_description: 待查询历史信息的起始时间，格式为 "2006-01-02 15:04:05"。
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: The end time for querying historical messages, formatted as "2006-01-02 15:04:05".
+      zh_Hans: 待查询历史信息的结束时间，格式为 "2006-01-02 15:04:05"。
+    llm_description: 待查询历史信息的结束时间，格式为 "2006-01-02 15:04:05"。
+    form: llm
+
+  - name: sort_type
+    type: select
+    required: false
+    options:
+      - value: ByCreateTimeAsc
+        label:
+          en_US: ByCreateTimeAsc
+          zh_Hans: ByCreateTimeAsc
+      - value: ByCreateTimeDesc
+        label:
+          en_US: ByCreateTimeDesc
+          zh_Hans: ByCreateTimeDesc
+    default: "ByCreateTimeAsc"
+    label:
+      en_US: Sort Type
+      zh_Hans: 排序方式
+    human_description:
+      en_US: |
+        The message sorting method. Optional values are ByCreateTimeAsc: sorted in ascending order by message creation time; ByCreateTimeDesc: sorted in descending order by message creation time. The default value is ByCreateTimeAsc. Note: When using page_token for pagination requests, the sorting method (sort_type) is consistent with the first request and cannot be changed midway.
+      zh_Hans: |
+        消息排序方式，可选值有 ByCreateTimeAsc：按消息创建时间升序排列；ByCreateTimeDesc：按消息创建时间降序排列。默认值为：ByCreateTimeAsc。注意：使用 page_token 分页请求时，排序方式（sort_type）均与第一次请求一致，不支持中途改换排序方式。
+    llm_description: 消息排序方式，可选值有 ByCreateTimeAsc：按消息创建时间升序排列；ByCreateTimeDesc：按消息创建时间降序排列。默认值为：ByCreateTimeAsc。注意：使用 page_token 分页请求时，排序方式（sort_type）均与第一次请求一致，不支持中途改换排序方式。
+    form: form
+
+  - name: page_size
+    type: number
+    required: false
+    default: 20
+    label:
+      en_US: Page Size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [1,50].
+      zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
+    llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: Page Token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: The pagination token. Leave it blank for the first request, indicating to start traversing from the beginning; when the pagination query result has more items, a new page_token will be returned simultaneously, which can be used to obtain the query result in the next traversal.
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.py b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.py
new file mode 100644
index 0000000000..3509d9bbcf
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetChatMessagesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        container_id = tool_parameters.get("container_id")
+        page_token = tool_parameters.get("page_token")
+        sort_type = tool_parameters.get("sort_type", "ByCreateTimeAsc")
+        page_size = tool_parameters.get("page_size", 20)
+
+        res = client.get_thread_messages(container_id, page_token, sort_type, page_size)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.yaml b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.yaml
new file mode 100644
index 0000000000..5f7a4f0902
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/get_thread_messages.yaml
@@ -0,0 +1,72 @@
+identity:
+  name: get_thread_messages
+  author: Doug Lea
+  label:
+    en_US: Get Thread Messages
+    zh_Hans: 获取指定话题的消息历史
+description:
+  human:
+    en_US: Get Thread Messages
+    zh_Hans: 获取指定话题的消息历史
+  llm: A tool for getting chat messages from specific threads.(获取指定话题的消息历史)
+parameters:
+  - name: container_id
+    type: string
+    required: true
+    label:
+      en_US: Thread Id
+      zh_Hans: 话题 ID
+    human_description:
+      en_US: The ID of the thread. Refer to the thread overview on how to obtain the thread_id. https://open.larksuite.com/document/uAjLw4CM/ukTMukTMukTM/reference/im-v1/message/thread-introduction
+      zh_Hans: 话题 ID，获取方式参见话题概述的如何获取 thread_id 章节。https://open.larksuite.com/document/uAjLw4CM/ukTMukTMukTM/reference/im-v1/message/thread-introduction
+    llm_description: 话题 ID，获取方式参见话题概述的如何获取 thread_id 章节。https://open.larksuite.com/document/uAjLw4CM/ukTMukTMukTM/reference/im-v1/message/thread-introduction
+    form: llm
+
+  - name: sort_type
+    type: select
+    required: false
+    options:
+      - value: ByCreateTimeAsc
+        label:
+          en_US: ByCreateTimeAsc
+          zh_Hans: ByCreateTimeAsc
+      - value: ByCreateTimeDesc
+        label:
+          en_US: ByCreateTimeDesc
+          zh_Hans: ByCreateTimeDesc
+    default: "ByCreateTimeAsc"
+    label:
+      en_US: Sort Type
+      zh_Hans: 排序方式
+    human_description:
+      en_US: |
+        The message sorting method. Optional values are ByCreateTimeAsc: sorted in ascending order by message creation time; ByCreateTimeDesc: sorted in descending order by message creation time. The default value is ByCreateTimeAsc. Note: When using page_token for pagination requests, the sorting method (sort_type) is consistent with the first request and cannot be changed midway.
+      zh_Hans: |
+        消息排序方式，可选值有 ByCreateTimeAsc：按消息创建时间升序排列；ByCreateTimeDesc：按消息创建时间降序排列。默认值为：ByCreateTimeAsc。注意：使用 page_token 分页请求时，排序方式（sort_type）均与第一次请求一致，不支持中途改换排序方式。
+    llm_description: 消息排序方式，可选值有 ByCreateTimeAsc：按消息创建时间升序排列；ByCreateTimeDesc：按消息创建时间降序排列。默认值为：ByCreateTimeAsc。注意：使用 page_token 分页请求时，排序方式（sort_type）均与第一次请求一致，不支持中途改换排序方式。
+    form: form
+
+  - name: page_size
+    type: number
+    required: false
+    default: 20
+    label:
+      en_US: Page Size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: The page size, i.e., the number of data entries returned in a single request. The default value is 20, and the value range is [1,50].
+      zh_Hans: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
+    llm_description: 分页大小，即单次请求所返回的数据条目数。默认值为 20，取值范围为 [1,50]。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: Page Token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: The pagination token. Leave it blank for the first request, indicating to start traversing from the beginning; when the pagination query result has more items, a new page_token will be returned simultaneously, which can be used to obtain the query result in the next traversal.
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.py b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.py
new file mode 100644
index 0000000000..b0a8df61e8
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class SendBotMessageTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        receive_id_type = tool_parameters.get("receive_id_type")
+        receive_id = tool_parameters.get("receive_id")
+        msg_type = tool_parameters.get("msg_type")
+        content = tool_parameters.get("content")
+
+        res = client.send_bot_message(receive_id_type, receive_id, msg_type, content)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.yaml b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.yaml
new file mode 100644
index 0000000000..b949c5e016
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_bot_message.yaml
@@ -0,0 +1,125 @@
+identity:
+  name: send_bot_message
+  author: Doug Lea
+  label:
+    en_US: Send Bot Message
+    zh_Hans: 发送 Lark 应用消息
+description:
+  human:
+    en_US: Send bot message
+    zh_Hans: 发送 Lark 应用消息
+  llm: A tool for sending Lark application messages.
+parameters:
+  - name: receive_id
+    type: string
+    required: true
+    label:
+      en_US: receive_id
+      zh_Hans: 消息接收者的 ID
+    human_description:
+      en_US: The ID of the message receiver, the ID type is consistent with the value of the query parameter receive_id_type.
+      zh_Hans: 消息接收者的 ID，ID 类型与查询参数 receive_id_type 的取值一致。
+    llm_description: 消息接收者的 ID，ID 类型与查询参数 receive_id_type 的取值一致。
+    form: llm
+
+  - name: receive_id_type
+    type: select
+    required: true
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+      - value: email
+        label:
+          en_US: email
+          zh_Hans: email
+      - value: chat_id
+        label:
+          en_US: chat_id
+          zh_Hans: chat_id
+    label:
+      en_US: receive_id_type
+      zh_Hans: 消息接收者的 ID 类型
+    human_description:
+      en_US: The ID type of the message receiver, optional values are open_id, union_id, user_id, email, chat_id, with a default value of open_id.
+      zh_Hans: 消息接收者的 ID 类型，可选值有 open_id、union_id、user_id、email、chat_id，默认值为 open_id。
+    llm_description: 消息接收者的 ID 类型，可选值有 open_id、union_id、user_id、email、chat_id，默认值为 open_id。
+    form: form
+
+  - name: msg_type
+    type: select
+    required: true
+    options:
+      - value: text
+        label:
+          en_US: text
+          zh_Hans: 文本
+      - value: interactive
+        label:
+          en_US: interactive
+          zh_Hans: 卡片
+      - value: post
+        label:
+          en_US: post
+          zh_Hans: 富文本
+      - value: image
+        label:
+          en_US: image
+          zh_Hans: 图片
+      - value: file
+        label:
+          en_US: file
+          zh_Hans: 文件
+      - value: audio
+        label:
+          en_US: audio
+          zh_Hans: 语音
+      - value: media
+        label:
+          en_US: media
+          zh_Hans: 视频
+      - value: sticker
+        label:
+          en_US: sticker
+          zh_Hans: 表情包
+      - value: share_chat
+        label:
+          en_US: share_chat
+          zh_Hans: 分享群名片
+      - value: share_user
+        label:
+          en_US: share_user
+          zh_Hans: 分享个人名片
+      - value: system
+        label:
+          en_US: system
+          zh_Hans: 系统消息
+    label:
+      en_US: msg_type
+      zh_Hans: 消息类型
+    human_description:
+      en_US: Message type. Optional values are text, post, image, file, audio, media, sticker, interactive, share_chat, share_user, system. For detailed introduction of different message types, refer to the message content(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json).
+      zh_Hans: 消息类型。可选值有：text、post、image、file、audio、media、sticker、interactive、share_chat、share_user、system。不同消息类型的详细介绍，参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    llm_description: 消息类型。可选值有：text、post、image、file、audio、media、sticker、interactive、share_chat、share_user、system。不同消息类型的详细介绍，参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    form: form
+
+  - name: content
+    type: string
+    required: true
+    label:
+      en_US: content
+      zh_Hans: 消息内容
+    human_description:
+      en_US: Message content, a JSON structure serialized string. The value of this parameter corresponds to msg_type. For example, if msg_type is text, this parameter needs to pass in text type content. To understand the format and usage limitations of different message types, refer to the message content(https://open.larksuite.com/document/server-docs/im-v1/message-content-description/create_json).
+      zh_Hans: 消息内容，JSON 结构序列化后的字符串。该参数的取值与 msg_type 对应，例如 msg_type 取值为 text，则该参数需要传入文本类型的内容。了解不同类型的消息内容格式、使用限制，可参见发送消息内容(https://open.larksuite.com/document/server-docs/im-v1/message-content-description/create_json)。
+    llm_description: 消息内容，JSON 结构序列化后的字符串。该参数的取值与 msg_type 对应，例如 msg_type 取值为 text，则该参数需要传入文本类型的内容。了解不同类型的消息内容格式、使用限制，可参见发送消息内容(https://open.larksuite.com/document/server-docs/im-v1/message-content-description/create_json)。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.py b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.py
new file mode 100644
index 0000000000..18a605079f
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class SendWebhookMessageTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        webhook = tool_parameters.get("webhook")
+        msg_type = tool_parameters.get("msg_type")
+        content = tool_parameters.get("content")
+
+        res = client.send_webhook_message(webhook, msg_type, content)
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.yaml b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.yaml
new file mode 100644
index 0000000000..ea13cae52b
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_message_and_group/tools/send_webhook_message.yaml
@@ -0,0 +1,68 @@
+identity:
+  name: send_webhook_message
+  author: Doug Lea
+  label:
+    en_US: Send Webhook Message
+    zh_Hans: 使用自定义机器人发送 Lark 消息
+description:
+  human:
+    en_US: Send webhook message
+    zh_Hans: 使用自定义机器人发送 Lark 消息
+  llm: A tool for sending Lark messages using a custom robot.
+parameters:
+  - name: webhook
+    type: string
+    required: true
+    label:
+      en_US: webhook
+      zh_Hans: webhook
+    human_description:
+      en_US: |
+        The address of the webhook, the format of the webhook address corresponding to the bot is as follows: https://open.larksuite.com/open-apis/bot/v2/hook/xxxxxxxxxxxxxxxxx. For details, please refer to: Lark Custom Bot Usage Guide(https://open.larkoffice.com/document/client-docs/bot-v3/add-custom-bot)
+      zh_Hans: |
+        webhook 的地址，机器人对应的 webhook 地址格式如下: https://open.larksuite.com/open-apis/bot/v2/hook/xxxxxxxxxxxxxxxxx，详情可参考: Lark 自定义机器人使用指南(https://open.larksuite.com/document/client-docs/bot-v3/add-custom-bot)
+    llm_description: |
+      webhook 的地址，机器人对应的 webhook 地址格式如下: https://open.larksuite.com/open-apis/bot/v2/hook/xxxxxxxxxxxxxxxxx，详情可参考: Lark 自定义机器人使用指南(https://open.larksuite.com/document/client-docs/bot-v3/add-custom-bot)
+    form: llm
+
+  - name: msg_type
+    type: select
+    required: true
+    options:
+      - value: text
+        label:
+          en_US: text
+          zh_Hans: 文本
+      - value: interactive
+        label:
+          en_US: interactive
+          zh_Hans: 卡片
+      - value: image
+        label:
+          en_US: image
+          zh_Hans: 图片
+      - value: share_chat
+        label:
+          en_US: share_chat
+          zh_Hans: 分享群名片
+    label:
+      en_US: msg_type
+      zh_Hans: 消息类型
+    human_description:
+      en_US: Message type. Optional values are text, image, interactive, share_chat. For detailed introduction of different message types, refer to the message content(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json).
+      zh_Hans: 消息类型。可选值有：text、image、interactive、share_chat。不同消息类型的详细介绍，参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    llm_description: 消息类型。可选值有：text、image、interactive、share_chat。不同消息类型的详细介绍，参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    form: form
+
+
+  - name: content
+    type: string
+    required: true
+    label:
+      en_US: content
+      zh_Hans: 消息内容
+    human_description:
+      en_US: Message content, a JSON structure serialized string. The value of this parameter corresponds to msg_type. For example, if msg_type is text, this parameter needs to pass in text type content. To understand the format and usage limitations of different message types, refer to the message content(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json).
+      zh_Hans: 消息内容，JSON 结构序列化后的字符串。该参数的取值与 msg_type 对应，例如 msg_type 取值为 text，则该参数需要传入文本类型的内容。了解不同类型的消息内容格式、使用限制，可参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    llm_description: 消息内容，JSON 结构序列化后的字符串。该参数的取值与 msg_type 对应，例如 msg_type 取值为 text，则该参数需要传入文本类型的内容。了解不同类型的消息内容格式、使用限制，可参见发送消息内容(https://open.larkoffice.com/document/server-docs/im-v1/message-content-description/create_json)。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/_assets/icon.png b/api/core/tools/provider/builtin/lark_spreadsheet/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..258b361261d4e3366251613141efaf200cd492db
GIT binary patch
literal 5417
zcmd^DS5%YRwoZN$O6VAxB7`D>B2{_|NCfFcC5m9^B1Ni1YN8;ag8~B5K}5ljNR=At
zDowYWE@T6ufDnpw#PfU4%XzwQcZ_?-eMtT>zS-xRYp%IgytSn<C*n8)0)cQ|GBL6R
zziWSftjyrQ-j&5i;0GFHYitN14~i^8Ajr8(Mh2JhPAmE08OXo$d)phDo_jX22sv$J
zzcoQ8X({$I4M1mYUXN33N<B-3jydw#j5@xBBVluXSp)LMJp7KQ;+1ZXbQwJu*A>sL
zdRI`?%X-mSlvNKO|Ie?xLgnG*o=vL-$74~@Z_^H692>&gwz#AF8KDeDRMbw;%S7@^
z0lEPo#d95ysLibL=R-lQaee?Fs}d9;GP^(x<}0F2F-h^b07#9l0oO$-H1Q3K391Vw
z<4=0p`{@z^2;xpdlh~~3JnE^>9z);@q9`KIu_BtdpQzbsD575%4Twvli5I4HvMd2)
zb~piMXifm)8t=*F!(<e9AZ#(&5I9%BCRU5h8n+7lA2M9bi$6TmHEA~ID(;`3wwo8D
z!HrgO5b^+&;dDiUJ{z6D?pevSmy#Cf>(F(ty-{~Wz)Z}Ndy@Eb!#j;_(=_NxxD1SZ
z`^;W2^5auNzYn@A2>xJh;sP2HSC+;woK59g(<;NCo^!c#Ehuot?D4Rd)w8!(U0P`&
z%VC9Q-96X{A0A-3$C;#tp*U^SO+lV&Hxa7rN-a66lFyi_>RrN%R|1Vq#cqxM66~=V
zqiED^xRL<xd16`2lz17QKLy`C;C5AQq}G?csiJ71;X1vkeAk?f3PH~lK?0n-*-Rh@
zP#+}EP<KhXST$KEuy5ja(B<qpz+U~Q@K@4cocJ1spl?)AdL-CEDd_Vhj4=~WZC!a~
zCD2iFc@$SoTjqPB#rFZiVeiN>X95ANScBy+VLmy`yfkVWj>n$$_jO&|;@RMZxW`Vz
z$O_A_ZC28OHeIiVe5Q^a@Eh@@xS_-0ekSH|OjbBisUizBYwacbqfha(|5(|VJ2}WB
zHpATzRjsPoJQBe4PIt1hgrI7Y<u3`Nkh4!riJ3M^ie0e!&bM~m=Y9#23n-n<2bDN3
z#0|w)^M+00X~2BYNeQ*hftMbP?ScCmGX4g&>50}-!}5gFUy?Rg9_E&^py-is{-p;%
zTvMdnGy%=Wqn{@HrmR@{jrI>dJKTelw2a|nx8U=k7~^*xb#q+marab_8vS2fSE5+Y
zn~5{LDooF9tC$SGL27*G@;{=nn$Xky=U;8P+72;ajTwrUrC1Tnxpj$-`QP(e{CvV9
zh3pN9onxvac@}^z>)z{0S+aM9#p?S7+MZA%@a=)}H%cBs2SJLPD0zo*4!;C}*GG6$
zTMl`6(oQ&vd#E0UZeZkChdQ~D8u|whYR0Tkk_54k46T{#q1zuLx1St1$09Z4@?G-x
zP;12H+5MiS`8ypOtP=3S-g>%7u{bLB#@8}4dCm`<qDYPN4vDb~$%Ks25&O8gW@kiR
zgX7!yjtl6cwxQzB^NKCexj>w!J@u0gpTDCfl`36#Ce`k^1%{YRQF_6`gSF9DyZ3Bm
zxr|mHO4@>BaQa#*$Cj2S)LE_&-nSsJv3#y!5=1E}zv^mBz1y3!vXp@@&b|hGYttj^
z?UFMVQR53YFP6~^{CEd_wTWBGlH@xTzP_*PtPwumo>==enAF|ol)S&NnfqRA-FZ#G
zf5A^UVYdF1#whcR86TO2%c|Bm2kp)#9~26{2Ge@<yY{ZgBQ<1jASZIHhfI<7W@U|^
z36W4PebMAzs*1_CW`);N&sM#Y2E2{p@~C%jETyXPhg1n5d|d6Zk1?=$Y4Lnv?@EVg
zZyTFlvtpXs-B{IeBV0RA;QqwR&w)=9W`)Ok{3z1wY$O3$rPku6tzH3vUiHASG31CF
zFEpM*#Jsxt^=6i7R80M?Y6{Yftufo?=HnY4t|9++sF~Fo_bckNnpAn`Tgm;LeiTs-
zY<c@=-(wo987{Xz=8%j{w*>9z!sOtO3(_ASdG)a5Jouh*GPceIQ8i;V2@|!6kT2)T
zP3G<Xwy01}Q<1t+HYJtfksF{91YtDzb}Vdj;%l}zYnzNa+QUGR3zIOi`DbURJzM<m
zja@guSjBiaPm`3txU(Mi!Srz9fhYC3YdD^2gZnjYgNyX2cQ-K!?mNF7@}o<EOE1Nf
zD}vi=+|MEB%jgASvuAK13_jxN5#F<WPT`*;$50oCFxN)%V_C}4NC|Dfzax#R(#mPI
z>TrKXy&4^~U@c_IEJKMZl%eR=P~7mddg1f~Y}6Wy#^*OPN3UfmQK^3@itZpKbd5@0
zc3B?kLSn(>gN`_VSnG<!@3hnH;|l8!a<hTuJLeEpLvl&N(>c2%ub#NtO@@Vkcuw1I
zE&gLUmn!~7K;3W8aMa%NswfiMeh!H(yq2vZu#y-1l|`&3hz-OPVp0Cr>%V@TkU!%>
zig$7Eo6aFiCaPc7AQg38k{ODANjtgts?q(!v%Aw&3~?*`#9gB9y+Zv%yYux6M}lWg
zZx<xJi+nC~VS409zF@4uNVBb(Mk2Z3f?3ZToAWoQPPAG2xhnrv;b?lb-g&$Lj^C$s
zPQ1~y0#}&SDev!H4o?gn=wXkGE^o&VB)#zL@8~aP&#cLPRP8*m&~BSz&nCyG#6jZf
zso1^MFWe;5nPB{~>}2Y?UvriIKFi_e_K9Y{*4iad-AWN{so#fZ7hA_A!jW=?o>IKy
zVmr>iln9}Jc@{D>cn$kyL;F=r9*Z3=5=?1+js+lA8D{cPQAguYGbtg}#)-6p{L3|*
z@z$zI?_#(<*R6L3#H|ws#J_XKWhkN21!JDd0#QqszIwaZd#vflMa8h(9b$wT5L8*<
zBX{8NtDYXEG^ec#34>iz2?zd{3O{5DZf>LtQt;n=e$*{$Wdmn-7<kklzDioRrO1qi
ztD2v1M8k&5NPXM=ZtXk0S@jJWL4l$s#4TPsT!=?KLJsBc{`*C?Lsa3$PS^6|JGhU7
z<9O}Ai?~pu>EK!_{#(~HLVkHiwO`?8f|1cmvooXKl(DSNxarLW);a=FHJAic)+i^G
z(e}q=C|W8Y@LuK-u<Tk2rV9klmYN`{Qj>v76%eROl8~k7sr_FPLKx5)D|=j5MXweA
zsi4aJnHaAJ-_CgY`b`pY7?BvwX)Ah}o;`g;d}iT<<{|np=0<vHKqNO<$dOFni<#Ck
zwZf90(-K<F91F>|Z$nPE>jWATfMx145R;)u$P_vwvE1OY&0~oC274T(v@nP)MM42l
z#rob;2tLAOLgc?^gTq+sb;?jg(TJ+PMBwcG4dIK#Er=}+<AlWWW(+AJs#wy1vl}I}
zgrbOtHaJ}Z5}PgvzY6=Rk_0S6%l>;)jzkBQbA6MY987*4wFXPvD(6q#>};6+{c!gE
zuh?~G=IFzw59PEky!+o7(m;57&_TyPg|dS<(VkwmfpJ;)jR3<_=3?xKgTG9sg|mT8
zmbR_XHYesG&?uhyDlEiy%>7r}!@aL&dpZ>1HoqDf{|1sCM-k|%3`JO`Qx1$Y1q`W#
z=265^17<`61QLL!MCE=k9QvOWA$*wMEB*<)^nabuVB#n)s@pcn@<vexczgsf$eP%I
z@3JN)&qV%0wWW7v<hh8>^cAx%#VAfr_bI)ekLa(&LqeVj$|@{;Y-L}+dGMs<i-qjB
zY?v+X*zMz*NG#KcadADHPGbeIc_wRiR__VO`E21E)ooPd$Kn_{&y~-FXi?!cZ;j)`
zW+86OO8FlQV|FKxW@02IOxZW46GL_i{b-fv6gpxnGhF$M{)3r=^@jeM+LX?&{Lh4+
zHcrFkpPy#et=tCH7Uz|oFd>G-d-d4iEN~E?1J;#SxCn*<otzM7&3l4eSClz$*8lga
zHdVb<<+LYwGXBl=3kMz!{TGQW{8#AugApQit$Hlb*59<5={EQdBR9^%cR7F(q@d+?
zUF^M=vW)+TbwL;kQN=xWk=T0WhWeVCx!7w+w`KfxNsITadP-Xo=L=PhVJeGo@a@!q
zmof3Jr(FYQWZ%24n76YqqWd-jE}xdBZyen;Hm_KpwW|Ccf?=F5QLnGQ<p3A84VD%?
zx#!K1n7me}#|Q^IXhd>=oaRy&&#~_7uBTMaU0*MMYxCUXn$sPyZ5ihR%j-)C1<G`f
z{%ZU%>apw>-l9JRzB?mJ5v?YFm*DN3ll)!1?9R}6Z3b@E^)gymONw#o8R29`3SRVA
zsi#m1t=0|{3Ol$DlC?xn@?&-b+n4<(hog<Jm9_sPW3HV8fsafcLt;}$drNc!+O^ii
zn#l17#w>e6@h^J8p`^7HP29r&!qo2Y?k_>r=O&21`&6h_Sx#E;kq2Ot#^)<fXdjDi
z`++b&abECiZm5^L^ZxzgpuM`se{A~JhFa*f?21g;L@CJs2)UGXaphhosJ6fkr|b0P
zVz~Ncyx8;V<NlJ&{F$XDNbi}u<112d5zuI^8K_;>I6TxR^2Lm5BW3iP4X-x4dph*G
zm6EL0WgHQStu6)~e{wh2tBkuxN}e`;DpR8pkKD&Fm)Mta1e}IKjl)|Et;Qyf)CwjV
zkPeeGRL>U?*#_-`_0#OH+8^nA;z!>_N-_;=+cGoY(3B$D=}mL@>(pjfZCk%evCrcJ
z&uIHP?BLXa=%?HV6Lp6PoN{`SM8K5p<G?dQ#bcfX!YmHW@&W8O<Q@<vPR##&j$>x4
z06A{wSth;olH1c3?@~tp92xRpW^+pix#O-i$m?O6I<%QK@99QH^!G_a5mi<-Wa`gE
zzEx$`fchvk6%$k<3><;_n^f0_-b<{r5(4(e|ICEo-FaWJS52Vs7-pVVc5U=O(B%;n
zv7{^76nDI(5CdMo4L}z>=DnZt{YJ#!;~_mwrpfX30Z_0@QJch%LX1fVD8<h{FpqS!
z0JzL5B-O2Ft{9^u80}UKZ#D+FhC&kuP?md|5)OKgclinJ|AbJ4FL<~UA#k-g>|}N$
z9N>nSWT$qR^Q{>WfTx|6jt&uOCFxfnj6UV(edaeo{}m!Cd`%DeX#7VPoz>WcK<6LQ
z^-L@MNLKzL>o41FvMqpzAe(q&>L>YY@!!*<)@KY!myRyFfXqBaH0EJT(-*!;(fo!w
zbJ0>MXdoA4mZsRnjJ{g5U${6KbuK2G9(^7p90v)^Re!~rYW$rXub|1*azcCG!`X<-
z0JsfOjz3>5x!8VHt5HK<(3JiZa<lHQlL|+=X5foM7FwQQx0%`HX0lgA&^T>Y8YsjE
z4#P-|aP9d+&hJ+~<z#dB<q*DSwx(<HvE@C4zz0`HDqdJ+Xxr@7X3V-wOXfS)i>x3)
zRoe}iD|SBMZC((X;TavAVgj57V|X*prE0P64W*hxeg&y88J~hekQ$+rMdeC*R;&Sn
zgx6C89qHlcp@_WZ?QnH`UrBv4bP7TGG&Rtb9xeyMog-$X+=pG(yy-#ELtWJFls{L%
zPG7OCf6C<aseo!CDL9J7-;W0!km5PNrW+4BXw}eLb#_@Wb1@Fq^*3b3Ek|rg7bk$o
zYnM*FsVX<^6xP=)U4WqX9}e{AAhJM#Xm9VOIr~IXHcQcQ^!eZX{yq1_(ZtP}_p*HH
z$rh*PBzaWmq_g5SkA9!ga5Vsfg4Wq|ytpo?pa76Fc|svVeAcJ()LLM<;@wI6OII06
zSC>;n*PmYab_Xq25TBgp_tTPf3g;+@O>2dfO)Z!zl~Sem?k8imS&3FK{d%ffu4JAj
zAjB0MS+yF#;WI_b8j*swkZ%2~-+wRp)VGRi>^J?MGjd<>nRdAr7r7&Dh`-Pv-FK>h
zIQo;wPAv-XS-wED56lI^%4ogV%jy{%p)hIaAjY;Y9uVr$e&h4y<vp@li5+4+XwhTU
zugYsO>)Uwd^W~3#Sm}JL6e>&pB)4%2Uz#8r-4KnR5f_ww?1ls1cbrIlQ*~F)F~(XH
z5fO%=Ejw1~Pfjtv;bPV{EI26c+pUV(<q*48d+&s2jam-S^~thP|Au<~N~=#wULsXy
z&8s4;J*m-E$Yl}RjtviJl6n*{6c|&%)b+-&NfN@50Tl9~_0elAeL+<1&GY533Fw|0
z><A@QAn#|P`l^6%xuASUl@q=loBB?Q2P_2<x#&SlpbgxtH+Qqmfw$mxdb{!MhZH>#
e^8bzLQ2*ztuUWBv%zNP5C*%^w(uiz`yZ<i?3E!pw

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.py b/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.py
new file mode 100644
index 0000000000..c791363f21
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkMessageProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.yaml
new file mode 100644
index 0000000000..030b5c9063
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/lark_spreadsheet.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_spreadsheet
+  label:
+    en_US: Lark Spreadsheet
+    zh_Hans: Lark 电子表格
+  description:
+    en_US: |
+      Lark Spreadsheet, requires the following permissions: sheets:spreadsheet.
+    zh_Hans: |
+      Lark 电子表格，需要开通以下权限: sheets:spreadsheet。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.py
new file mode 100644
index 0000000000..deeb5a1ecf
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.py
@@ -0,0 +1,22 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class AddColsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        sheet_id = tool_parameters.get("sheet_id")
+        sheet_name = tool_parameters.get("sheet_name")
+        length = tool_parameters.get("length")
+        values = tool_parameters.get("values")
+
+        res = client.add_cols(spreadsheet_token, sheet_id, sheet_name, length, values)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.yaml
new file mode 100644
index 0000000000..b73335f405
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_cols.yaml
@@ -0,0 +1,72 @@
+identity:
+  name: add_cols
+  author: Doug Lea
+  label:
+    en_US: Add Cols
+    zh_Hans: 新增多列至工作表最后
+description:
+  human:
+    en_US: Add Cols
+    zh_Hans: 新增多列至工作表最后
+  llm: A tool for adding multiple columns to the end of a spreadsheet. (新增多列至工作表最后)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: spreadsheet_token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 url。
+    llm_description: 电子表格 token，支持输入电子表格 url。
+    form: llm
+
+  - name: sheet_id
+    type: string
+    required: false
+    label:
+      en_US: sheet_id
+      zh_Hans: 工作表 ID
+    human_description:
+      en_US: Sheet ID, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表 ID，与 sheet_name 二者其一必填。
+    llm_description: 工作表 ID，与 sheet_name 二者其一必填。
+    form: llm
+
+  - name: sheet_name
+    type: string
+    required: false
+    label:
+      en_US: sheet_name
+      zh_Hans: 工作表名称
+    human_description:
+      en_US: Sheet name, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表名称，与 sheet_id 二者其一必填。
+    llm_description: 工作表名称，与 sheet_id 二者其一必填。
+    form: llm
+
+  - name: length
+    type: number
+    required: true
+    label:
+      en_US: length
+      zh_Hans: 要增加的列数
+    human_description:
+      en_US: Number of columns to add, range (0-5000].
+      zh_Hans: 要增加的列数，范围(0-5000]。
+    llm_description: 要增加的列数，范围(0-5000]。
+    form: form
+
+  - name: values
+    type: string
+    required: false
+    label:
+      en_US: values
+      zh_Hans: 新增列的单元格内容
+    human_description:
+      en_US: |
+        Content of the new columns, array of objects in string format, each array represents a row of table data, format like: [ [ "ID","Name","Age" ],[ 1,"Zhang San",10 ],[ 2,"Li Si",11 ] ].
+      zh_Hans: 新增列的单元格内容，数组对象字符串，每个数组一行表格数据，格式：[["编号","姓名","年龄"],[1,"张三",10],[2,"李四",11]]。
+    llm_description: 新增列的单元格内容，数组对象字符串，每个数组一行表格数据，格式：[["编号","姓名","年龄"],[1,"张三",10],[2,"李四",11]]。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.py
new file mode 100644
index 0000000000..f434b1c603
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.py
@@ -0,0 +1,22 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class AddRowsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        sheet_id = tool_parameters.get("sheet_id")
+        sheet_name = tool_parameters.get("sheet_name")
+        length = tool_parameters.get("length")
+        values = tool_parameters.get("values")
+
+        res = client.add_rows(spreadsheet_token, sheet_id, sheet_name, length, values)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.yaml
new file mode 100644
index 0000000000..6bce305b98
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/add_rows.yaml
@@ -0,0 +1,72 @@
+identity:
+  name: add_rows
+  author: Doug Lea
+  label:
+    en_US: Add Rows
+    zh_Hans: 新增多行至工作表最后
+description:
+  human:
+    en_US: Add Rows
+    zh_Hans: 新增多行至工作表最后
+  llm: A tool for adding multiple rows to the end of a spreadsheet. (新增多行至工作表最后)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: spreadsheet_token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 url。
+    llm_description: 电子表格 token，支持输入电子表格 url。
+    form: llm
+
+  - name: sheet_id
+    type: string
+    required: false
+    label:
+      en_US: sheet_id
+      zh_Hans: 工作表 ID
+    human_description:
+      en_US: Sheet ID, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表 ID，与 sheet_name 二者其一必填。
+    llm_description: 工作表 ID，与 sheet_name 二者其一必填。
+    form: llm
+
+  - name: sheet_name
+    type: string
+    required: false
+    label:
+      en_US: sheet_name
+      zh_Hans: 工作表名称
+    human_description:
+      en_US: Sheet name, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表名称，与 sheet_id 二者其一必填。
+    llm_description: 工作表名称，与 sheet_id 二者其一必填。
+    form: llm
+
+  - name: length
+    type: number
+    required: true
+    label:
+      en_US: length
+      zh_Hans: 要增加行数
+    human_description:
+      en_US: Number of rows to add, range (0-5000].
+      zh_Hans: 要增加行数，范围(0-5000]。
+    llm_description: 要增加行数，范围(0-5000]。
+    form: form
+
+  - name: values
+    type: string
+    required: false
+    label:
+      en_US: values
+      zh_Hans: 新增行的表格内容
+    human_description:
+      en_US: |
+        Content of the new rows, array of objects in string format, each array represents a row of table data, format like: [ [ "ID","Name","Age" ],[ 1,"Zhang San",10 ],[ 2,"Li Si",11 ] ].
+      zh_Hans: 新增行的表格内容，数组对象字符串，每个数组一行表格数据，格式,如：[["编号","姓名","年龄"],[1,"张三",10],[2,"李四",11]]。
+    llm_description: 新增行的表格内容，数组对象字符串，每个数组一行表格数据，格式,如：[["编号","姓名","年龄"],[1,"张三",10],[2,"李四",11]]。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.py
new file mode 100644
index 0000000000..74b20ac2c8
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateSpreadsheetTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        title = tool_parameters.get("title")
+        folder_token = tool_parameters.get("folder_token")
+
+        res = client.create_spreadsheet(title, folder_token)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.yaml
new file mode 100644
index 0000000000..931310e631
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/create_spreadsheet.yaml
@@ -0,0 +1,35 @@
+identity:
+  name: create_spreadsheet
+  author: Doug Lea
+  label:
+    en_US: Create Spreadsheet
+    zh_Hans: 创建电子表格
+description:
+  human:
+    en_US: Create Spreadsheet
+    zh_Hans: 创建电子表格
+  llm: A tool for creating spreadsheets. (创建电子表格)
+parameters:
+  - name: title
+    type: string
+    required: false
+    label:
+      en_US: Spreadsheet Title
+      zh_Hans: 电子表格标题
+    human_description:
+      en_US: The title of the spreadsheet
+      zh_Hans: 电子表格的标题
+    llm_description: 电子表格的标题
+    form: llm
+
+  - name: folder_token
+    type: string
+    required: false
+    label:
+      en_US: Folder Token
+      zh_Hans: 文件夹 token
+    human_description:
+      en_US: The token of the folder, supports folder URL input, e.g., https://bytedance.larkoffice.com/drive/folder/CxHEf4DCSlNkL2dUTCJcPRgentg
+      zh_Hans: 文件夹 token，支持文件夹 URL 输入，如：https://bytedance.larkoffice.com/drive/folder/CxHEf4DCSlNkL2dUTCJcPRgentg
+    llm_description: 文件夹 token，支持文件夹 URL 输入，如：https://bytedance.larkoffice.com/drive/folder/CxHEf4DCSlNkL2dUTCJcPRgentg
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.py
new file mode 100644
index 0000000000..0fe35b6dc6
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.py
@@ -0,0 +1,19 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetSpreadsheetTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.get_spreadsheet(spreadsheet_token, user_id_type)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.yaml
new file mode 100644
index 0000000000..c519938617
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/get_spreadsheet.yaml
@@ -0,0 +1,49 @@
+identity:
+  name: get_spreadsheet
+  author: Doug Lea
+  label:
+    en_US: Get Spreadsheet
+    zh_Hans: 获取电子表格信息
+description:
+  human:
+    en_US: Get Spreadsheet
+    zh_Hans: 获取电子表格信息
+  llm: A tool for getting information from spreadsheets. (获取电子表格信息)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: Spreadsheet Token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 URL。
+    llm_description: 电子表格 token，支持输入电子表格 URL。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.py
new file mode 100644
index 0000000000..e711c23780
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.py
@@ -0,0 +1,18 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ListSpreadsheetSheetsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+
+        res = client.list_spreadsheet_sheets(spreadsheet_token)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.yaml
new file mode 100644
index 0000000000..c6a7ef45d4
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/list_spreadsheet_sheets.yaml
@@ -0,0 +1,23 @@
+identity:
+  name: list_spreadsheet_sheets
+  author: Doug Lea
+  label:
+    en_US: List Spreadsheet Sheets
+    zh_Hans: 列出电子表格所有工作表
+description:
+  human:
+    en_US: List Spreadsheet Sheets
+    zh_Hans: 列出电子表格所有工作表
+  llm: A tool for listing all sheets in a spreadsheet. (列出电子表格所有工作表)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: Spreadsheet Token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 URL。
+    llm_description: 电子表格 token，支持输入电子表格 URL。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.py
new file mode 100644
index 0000000000..1df289c1d7
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ReadColsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        sheet_id = tool_parameters.get("sheet_id")
+        sheet_name = tool_parameters.get("sheet_name")
+        start_col = tool_parameters.get("start_col")
+        num_cols = tool_parameters.get("num_cols")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.read_cols(spreadsheet_token, sheet_id, sheet_name, start_col, num_cols, user_id_type)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.yaml
new file mode 100644
index 0000000000..34da74592d
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_cols.yaml
@@ -0,0 +1,97 @@
+identity:
+  name: read_cols
+  author: Doug Lea
+  label:
+    en_US: Read Cols
+    zh_Hans: 读取工作表列数据
+description:
+  human:
+    en_US: Read Cols
+    zh_Hans: 读取工作表列数据
+  llm: A tool for reading column data from a spreadsheet. (读取工作表列数据)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: spreadsheet_token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 url。
+    llm_description: 电子表格 token，支持输入电子表格 url。
+    form: llm
+
+  - name: sheet_id
+    type: string
+    required: false
+    label:
+      en_US: sheet_id
+      zh_Hans: 工作表 ID
+    human_description:
+      en_US: Sheet ID, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表 ID，与 sheet_name 二者其一必填。
+    llm_description: 工作表 ID，与 sheet_name 二者其一必填。
+    form: llm
+
+  - name: sheet_name
+    type: string
+    required: false
+    label:
+      en_US: sheet_name
+      zh_Hans: 工作表名称
+    human_description:
+      en_US: Sheet name, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表名称，与 sheet_id 二者其一必填。
+    llm_description: 工作表名称，与 sheet_id 二者其一必填。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: start_col
+    type: number
+    required: false
+    label:
+      en_US: start_col
+      zh_Hans: 起始列号
+    human_description:
+      en_US: Starting column number, starting from 1.
+      zh_Hans: 起始列号，从 1 开始。
+    llm_description: 起始列号，从 1 开始。
+    form: form
+
+  - name: num_cols
+    type: number
+    required: true
+    label:
+      en_US: num_cols
+      zh_Hans: 读取列数
+    human_description:
+      en_US: Number of columns to read.
+      zh_Hans: 读取列数
+    llm_description: 读取列数
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.py
new file mode 100644
index 0000000000..1cab38a454
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ReadRowsTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        sheet_id = tool_parameters.get("sheet_id")
+        sheet_name = tool_parameters.get("sheet_name")
+        start_row = tool_parameters.get("start_row")
+        num_rows = tool_parameters.get("num_rows")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.read_rows(spreadsheet_token, sheet_id, sheet_name, start_row, num_rows, user_id_type)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.yaml
new file mode 100644
index 0000000000..5dfa8d5835
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_rows.yaml
@@ -0,0 +1,97 @@
+identity:
+  name: read_rows
+  author: Doug Lea
+  label:
+    en_US: Read Rows
+    zh_Hans: 读取工作表行数据
+description:
+  human:
+    en_US: Read Rows
+    zh_Hans: 读取工作表行数据
+  llm: A tool for reading row data from a spreadsheet. (读取工作表行数据)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: spreadsheet_token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 url。
+    llm_description: 电子表格 token，支持输入电子表格 url。
+    form: llm
+
+  - name: sheet_id
+    type: string
+    required: false
+    label:
+      en_US: sheet_id
+      zh_Hans: 工作表 ID
+    human_description:
+      en_US: Sheet ID, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表 ID，与 sheet_name 二者其一必填。
+    llm_description: 工作表 ID，与 sheet_name 二者其一必填。
+    form: llm
+
+  - name: sheet_name
+    type: string
+    required: false
+    label:
+      en_US: sheet_name
+      zh_Hans: 工作表名称
+    human_description:
+      en_US: Sheet name, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表名称，与 sheet_id 二者其一必填。
+    llm_description: 工作表名称，与 sheet_id 二者其一必填。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: start_row
+    type: number
+    required: false
+    label:
+      en_US: start_row
+      zh_Hans: 起始行号
+    human_description:
+      en_US: Starting row number, starting from 1.
+      zh_Hans: 起始行号，从 1 开始。
+    llm_description: 起始行号，从 1 开始。
+    form: form
+
+  - name: num_rows
+    type: number
+    required: true
+    label:
+      en_US: num_rows
+      zh_Hans: 读取行数
+    human_description:
+      en_US: Number of rows to read.
+      zh_Hans: 读取行数
+    llm_description: 读取行数
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.py b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.py
new file mode 100644
index 0000000000..0f05249004
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class ReadTableTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        spreadsheet_token = tool_parameters.get("spreadsheet_token")
+        sheet_id = tool_parameters.get("sheet_id")
+        sheet_name = tool_parameters.get("sheet_name")
+        num_range = tool_parameters.get("num_range")
+        query = tool_parameters.get("query")
+        user_id_type = tool_parameters.get("user_id_type", "open_id")
+
+        res = client.read_table(spreadsheet_token, sheet_id, sheet_name, num_range, query, user_id_type)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.yaml b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.yaml
new file mode 100644
index 0000000000..10534436d6
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_spreadsheet/tools/read_table.yaml
@@ -0,0 +1,122 @@
+identity:
+  name: read_table
+  author: Doug Lea
+  label:
+    en_US: Read Table
+    zh_Hans: 自定义读取电子表格行列数据
+description:
+  human:
+    en_US: Read Table
+    zh_Hans: 自定义读取电子表格行列数据
+  llm: A tool for custom reading of row and column data from a spreadsheet. (自定义读取电子表格行列数据)
+parameters:
+  - name: spreadsheet_token
+    type: string
+    required: true
+    label:
+      en_US: spreadsheet_token
+      zh_Hans: 电子表格 token
+    human_description:
+      en_US: Spreadsheet token, supports input of spreadsheet URL.
+      zh_Hans: 电子表格 token，支持输入电子表格 url。
+    llm_description: 电子表格 token，支持输入电子表格 url。
+    form: llm
+
+  - name: sheet_id
+    type: string
+    required: false
+    label:
+      en_US: sheet_id
+      zh_Hans: 工作表 ID
+    human_description:
+      en_US: Sheet ID, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表 ID，与 sheet_name 二者其一必填。
+    llm_description: 工作表 ID，与 sheet_name 二者其一必填。
+    form: llm
+
+  - name: sheet_name
+    type: string
+    required: false
+    label:
+      en_US: sheet_name
+      zh_Hans: 工作表名称
+    human_description:
+      en_US: Sheet name, either sheet_id or sheet_name must be filled.
+      zh_Hans: 工作表名称，与 sheet_id 二者其一必填。
+    llm_description: 工作表名称，与 sheet_id 二者其一必填。
+    form: llm
+
+  - name: user_id_type
+    type: select
+    required: false
+    options:
+      - value: open_id
+        label:
+          en_US: open_id
+          zh_Hans: open_id
+      - value: union_id
+        label:
+          en_US: union_id
+          zh_Hans: union_id
+      - value: user_id
+        label:
+          en_US: user_id
+          zh_Hans: user_id
+    default: "open_id"
+    label:
+      en_US: user_id_type
+      zh_Hans: 用户 ID 类型
+    human_description:
+      en_US: User ID type, optional values are open_id, union_id, user_id, with a default value of open_id.
+      zh_Hans: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    llm_description: 用户 ID 类型，可选值有 open_id、union_id、user_id，默认值为 open_id。
+    form: form
+
+  - name: start_row
+    type: number
+    required: false
+    label:
+      en_US: start_row
+      zh_Hans: 起始行号
+    human_description:
+      en_US: Starting row number, starting from 1.
+      zh_Hans: 起始行号，从 1 开始。
+    llm_description: 起始行号，从 1 开始。
+    form: form
+
+  - name: num_rows
+    type: number
+    required: false
+    label:
+      en_US: num_rows
+      zh_Hans: 读取行数
+    human_description:
+      en_US: Number of rows to read.
+      zh_Hans: 读取行数
+    llm_description: 读取行数
+    form: form
+
+  - name: range
+    type: string
+    required: false
+    label:
+      en_US: range
+      zh_Hans: 取数范围
+    human_description:
+      en_US: |
+        Data range, format like: A1:B2, can be empty when query=all.
+      zh_Hans: 取数范围,格式如:A1:B2，query=all 时可为空。
+    llm_description: 取数范围,格式如:A1:B2，query=all 时可为空。
+    form: llm
+
+  - name: query
+    type: string
+    required: false
+    label:
+      en_US: query
+      zh_Hans: 查询
+    human_description:
+      en_US: Pass "all" to query all data in the table, but no more than 100 columns.
+      zh_Hans: 传 all，表示查询表格所有数据，但最多查询 100 列数据。
+    llm_description: 传 all，表示查询表格所有数据，但最多查询 100 列数据。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_task/_assets/icon.png b/api/core/tools/provider/builtin/lark_task/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..26ea6a2eefa5bea073ed8ca8c4d712852745026c
GIT binary patch
literal 2115
zcmV-J2)y@+P)<h;3K|Lk000e1NJLTq002M$002M;1^@s6s%dfF000OINkl<ZcwX(7
z+fox*6o!2SAAv{VQUx{2Asi$f3`c<gata`dqBWPp8|WQ6sOX^AUQUgO8GHazH3vZq
zL;=lt?^*xad#Bq51Q8vR=~eYtqf*5G^I!jJj~(;I8*jYv#v5<E@x~i(yz!>ECvr`e
zMW$rA`}a$J$awka4G|l^DdqT0A+Ow$%U6Gr*WBM%k~4Z!T9NBQKm<XIO$mm;AW9v=
z@YS3(GMST+>v>2-Zsg@C$v~halrDq;GCy=RCm=u=nat}H0D_QXoB}CIG6^t0IFXfP
z9umU@02;Zj(x4&eb)Xd7Dj^8{<2mc{m5dx90m7h4fgHYu6euYO+zFh7u_+{>1lFN{
zJY)5br6oOufiN(U(GapK1(SJ70y;sj!xR9RBor}(@My;B9Zk!=(X{9rBLL$W2?&r{
z2VMz*Bp47n2^s>YK#T)IvArvVhci|<l9J&_TJ(;l2ttM+APH#(FmNR+1}Ac?p=;Ss
z*Wrc+QAh$`hhFz*>^t)d+4A|{vy)Gs=X>XVlV$Ed|5npQ=6i;dvS%bE0HHSm0F(p*
zfE4s24TVl12uK1_U?-t)9?gjyo61kSKWERGi^}6qo{MKsp68`gPZz%ZE_DrLtgfMt
zvKs(M;gOV5gx7!}U>!^+T&5%pOdt)q9LfAJ$x)1y1bWiTll!xF9c@19Hvc^!boM8$
zOM@RJ)IF4B2;IY!0tTRDlE5j@oiG6N48Tr;>pCpEKWo>4`S}GAonOey`R7{r+Wflk
zJbDN)-*Nea>?B<p0EDFI0)!!zgdTw4K1fUM1a%p>6OaaV9kP5G_&VryAOJD^hsphU
zJ4Tt$M}J+&$%ANqF@IzbV7|Tog9HGEKmZ^OL8v6?PM|dC%fQ8m6RDCwtH8ymSAmm|
zUp}z;$1lkI3o_@@V;IJB7aawJws6vF?foFy`nF|T-v`mr|4}Fi3_@TXkOW8<BXt5L
z0p+NZV9L??vP1w5{x1*nZgU6GqpySz3U6C2;Vs$P3n>6ldjSF`fg#wPV6TH&2`)t^
zgo#Y7FOn>Cf8MV1=w<HNqeU|F6`EVj^tpfTegpHNo-GmT*_18eZAbtD(tvd!39}0J
zI;8A%;Bs_AFrDxP=I37Id~=Igd1$@VTFfSpB=^tXYnr<@Es|{R-VzrH06-`JDhaJR
z3HCaC)Wyi_pvqCNgL+pmFH7lzo5ws@Oi#kVL#@5OK9FAJ&boj8ZWAE$!AtLDu<O0N
z*uBX@3J8M4Isk;~gKbIyWEZ13k^G*(5K^LNM7=9e3YNdfJm%qIX1ZV?y8H8i`P!ui
z=85m2=Ug;f?*94i33P5+fzEf*>U<B0AX1={pjJUauscEZ0y;shg6l-;bszw-jxRNj
znWeup{_=;>eKLN(@atUc_C71n*qN}J+TRIC1CX4AOPiv(i&lY?uvJKc>H`)hLM1_g
z00h1agE*0?<wKb_GaF6cx&KE-9#*qa(qihfB?$8u+7c3C2p#WapyNG*Q0qVtx{(B|
z1k!*`FiF_b>!6Ae5Fp(Lhx?az4<O9K%pr6ynx1t3Jm=K6C9H;)4cX9|5RD8$0RV^q
zJyjZz0<{kAB<Sm4r@-!nSV!Lxm^U-KL=cuT@)r`GPMP1svmlsX+Q)n-VbzB=WIX_o
z8e2Cw1q`8`ARr0o1YQNR4$fla6De9g>4a@SI9l^&0O1pp4&B?K&zrApj$8hV>#{Di
zF6&tvLP1DC0uV0fbwCOND8&K<=Y`mD9mss5%>8*sHe~G1y;o_ur<T%k<^bkvo7bfu
zx(IEE9T0R100OIkP5=Ol0N_bU;9`XA>+p@uo0(ZkGK9Mh<K}Dnvq|^&JEt}fw`#1o
zthVB!Cb%vLgy19)ggUH)hJa4kQH+HX`A(h3E*nHBy!n01o4LD`nl`5Kyt|Z6(C>Y=
z?W@TA`KC28|CUt~jLTZ6InHUQ(<{-?k^lgm1cK16AOx@qyb>k_O7ylE3wC@*^JZqE
z$ueqr-^{CIaKAf$t}$+%yRatDLrrf5!4M3ARLe;Kgg8S$F>(r8=rTAu0V|=_0cqef
z2@wB{`7fC>^=p=|@r^v!XaE=pl!Sm;1th^<2X!41x)U}y2~8>q>N==$EDrN$>fZ=o
z!y5_NDOi&V08$Wltb>7obs!M>MDlw=t3Hn^2|vF1%DR>E)BaTf0MJ=L&`DqrCIzMw
zcpX$9_!S5wfz^N%v=&Nnx$Z|dk2zhp8msiLNP5aMPyvDhpdrYrCbbHj22KK**C$e6
z2K}<cZ;P>_Fn^-@ukwo86<JZcDk>=j458i$fs-H#>p*F!#wsX%+6f`^!n9Ew<}m<p
zss;d7BvPOuU=?f#4Qs+A;oJpNidYBT36ui94tQA#Z7kOXi^M#}=ZhUbS^Zivgqjsm
z$shm$eXuIe+Uvj&>?Fk1i6o&9a3TD`^;l7vH*>1$?`c4&Fc8oOe$xj8fdDums7~N@
zP{qg))|Y+7Xx_}p>XkbLK~NHeT8C9$2~~`gg8DUAF}@Xr6KTCA^DAZUpILNsqUxX9
zr)qc|Gz29ifmVUV?+JwyNqx{%BIeDIc@2S9qV}~)0i{6|qhlRVjE!$%C1u_WAn0Or
zB*9sX3?k;MD?#&SPE-Mc0^nQ+{jyZr=FL=8zy4XR!>R^BDe(VK%$qq`^{4M7ok=>6
tw=4Z?z2)@Y;Qk(Oyz#~xZ@l^b^Dp0fX>jNWgCYO`002ovPDHLkV1hdG!s7q{

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_task/lark_task.py b/api/core/tools/provider/builtin/lark_task/lark_task.py
new file mode 100644
index 0000000000..02cf009f01
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/lark_task.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkTaskProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_task/lark_task.yaml b/api/core/tools/provider/builtin/lark_task/lark_task.yaml
new file mode 100644
index 0000000000..ada068b0aa
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/lark_task.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_task
+  label:
+    en_US: Lark Task
+    zh_Hans: Lark 任务
+  description:
+    en_US: |
+      Lark Task, requires the following permissions: task:task:write、contact:user.id:readonly.
+    zh_Hans: |
+      Lark 任务，需要开通以下权限: task:task:write、contact:user.id:readonly。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_task/tools/add_members.py b/api/core/tools/provider/builtin/lark_task/tools/add_members.py
new file mode 100644
index 0000000000..9b8e4d68f3
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/add_members.py
@@ -0,0 +1,20 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class AddMembersTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        task_guid = tool_parameters.get("task_guid")
+        member_phone_or_email = tool_parameters.get("member_phone_or_email")
+        member_role = tool_parameters.get("member_role", "follower")
+
+        res = client.add_members(task_guid, member_phone_or_email, member_role)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_task/tools/add_members.yaml b/api/core/tools/provider/builtin/lark_task/tools/add_members.yaml
new file mode 100644
index 0000000000..0b12172e0b
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/add_members.yaml
@@ -0,0 +1,58 @@
+identity:
+  name: add_members
+  author: Doug Lea
+  label:
+    en_US: Add Lark Members
+    zh_Hans: 添加 Lark 任务成员
+description:
+  human:
+    en_US: Add Lark Members
+    zh_Hans: 添加 Lark 任务成员
+  llm: A tool for adding members to a Lark task.(添加 Lark 任务成员)
+parameters:
+  - name: task_guid
+    type: string
+    required: true
+    label:
+      en_US: Task GUID
+      zh_Hans: 任务 GUID
+    human_description:
+      en_US: |
+        The GUID of the task to be added, supports passing either the Task ID or the Task link URL. Example of Task ID: 8b5425ec-9f2a-43bd-a3ab-01912f50282b; Example of Task link URL: https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+      zh_Hans: 要添加的任务的 GUID，支持传任务 ID 和任务链接 URL。任务 ID 示例：8b5425ec-9f2a-43bd-a3ab-01912f50282b；任务链接 URL 示例：https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    llm_description: 要添加的任务的 GUID，支持传任务 ID 和任务链接 URL。任务 ID 示例：8b5425ec-9f2a-43bd-a3ab-01912f50282b；任务链接 URL 示例：https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    form: llm
+
+  - name: member_phone_or_email
+    type: string
+    required: true
+    label:
+      en_US: Task Member Phone Or Email
+      zh_Hans: 任务成员的电话或邮箱
+    human_description:
+      en_US: A list of member emails or phone numbers, separated by commas.
+      zh_Hans: 任务成员邮箱或者手机号列表，使用逗号分隔。
+    llm_description: 任务成员邮箱或者手机号列表，使用逗号分隔。
+    form: llm
+
+  - name: member_role
+    type: select
+    required: true
+    options:
+      - value: assignee
+        label:
+          en_US: assignee
+          zh_Hans: 负责人
+      - value: follower
+        label:
+          en_US: follower
+          zh_Hans: 关注人
+    default: "follower"
+    label:
+      en_US: member_role
+      zh_Hans: 成员的角色
+    human_description:
+      en_US: Member role, optional values are "assignee" (responsible person) and "follower" (observer), with a default value of "assignee".
+      zh_Hans: 成员的角色，可选值有 "assignee"（负责人）和 "follower"（关注人），默认值为 "assignee"。
+    llm_description: 成员的角色，可选值有 "assignee"（负责人）和 "follower"（关注人），默认值为 "assignee"。
+    form: form
diff --git a/api/core/tools/provider/builtin/lark_task/tools/create_task.py b/api/core/tools/provider/builtin/lark_task/tools/create_task.py
new file mode 100644
index 0000000000..ff37593fbe
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/create_task.py
@@ -0,0 +1,22 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class CreateTaskTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        summary = tool_parameters.get("summary")
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        completed_time = tool_parameters.get("completed_time")
+        description = tool_parameters.get("description")
+
+        res = client.create_task(summary, start_time, end_time, completed_time, description)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_task/tools/create_task.yaml b/api/core/tools/provider/builtin/lark_task/tools/create_task.yaml
new file mode 100644
index 0000000000..4303763a1d
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/create_task.yaml
@@ -0,0 +1,74 @@
+identity:
+  name: create_task
+  author: Doug Lea
+  label:
+    en_US: Create Lark Task
+    zh_Hans: 创建 Lark 任务
+description:
+  human:
+    en_US: Create Lark Task
+    zh_Hans: 创建 Lark 任务
+  llm: A tool for creating tasks in Lark.(创建 Lark 任务)
+parameters:
+  - name: summary
+    type: string
+    required: true
+    label:
+      en_US: Task Title
+      zh_Hans: 任务标题
+    human_description:
+      en_US: The title of the task.
+      zh_Hans: 任务标题
+    llm_description: 任务标题
+    form: llm
+
+  - name: description
+    type: string
+    required: false
+    label:
+      en_US: Task Description
+      zh_Hans: 任务备注
+    human_description:
+      en_US: The description or notes for the task.
+      zh_Hans: 任务备注
+    llm_description: 任务备注
+    form: llm
+
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 任务开始时间
+    human_description:
+      en_US: |
+        The start time of the task, in the format: 2006-01-02 15:04:05
+      zh_Hans: 任务开始时间，格式为：2006-01-02 15:04:05
+    llm_description: 任务开始时间，格式为：2006-01-02 15:04:05
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 任务结束时间
+    human_description:
+      en_US: |
+        The end time of the task, in the format: 2006-01-02 15:04:05
+      zh_Hans: 任务结束时间，格式为：2006-01-02 15:04:05
+    llm_description: 任务结束时间，格式为：2006-01-02 15:04:05
+    form: llm
+
+  - name: completed_time
+    type: string
+    required: false
+    label:
+      en_US: Completed Time
+      zh_Hans: 任务完成时间
+    human_description:
+      en_US: |
+        The completion time of the task, in the format: 2006-01-02 15:04:05. Leave empty to create an incomplete task; fill in a specific time to create a completed task.
+      zh_Hans: 任务完成时间，格式为：2006-01-02 15:04:05，不填写表示创建一个未完成任务；填写一个具体的时间表示创建一个已完成任务。
+    llm_description: 任务完成时间，格式为：2006-01-02 15:04:05，不填写表示创建一个未完成任务；填写一个具体的时间表示创建一个已完成任务。
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_task/tools/delete_task.py b/api/core/tools/provider/builtin/lark_task/tools/delete_task.py
new file mode 100644
index 0000000000..eca381be2c
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/delete_task.py
@@ -0,0 +1,18 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class UpdateTaskTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        task_guid = tool_parameters.get("task_guid")
+
+        res = client.delete_task(task_guid)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_task/tools/delete_task.yaml b/api/core/tools/provider/builtin/lark_task/tools/delete_task.yaml
new file mode 100644
index 0000000000..bc0154d9dc
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/delete_task.yaml
@@ -0,0 +1,24 @@
+identity:
+  name: delete_task
+  author: Doug Lea
+  label:
+    en_US: Delete Lark Task
+    zh_Hans: 删除 Lark 任务
+description:
+  human:
+    en_US: Delete Lark Task
+    zh_Hans: 删除 Lark 任务
+  llm: A tool for deleting tasks in Lark.(删除 Lark 任务)
+parameters:
+  - name: task_guid
+    type: string
+    required: true
+    label:
+      en_US: Task GUID
+      zh_Hans: 任务 GUID
+    human_description:
+      en_US: |
+        The GUID of the task to be deleted, supports passing either the Task ID or the Task link URL. Example of Task ID: 8b5425ec-9f2a-43bd-a3ab-01912f50282b; Example of Task link URL: https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+      zh_Hans: 要删除的任务的 GUID，支持传任务 ID 和任务链接 URL。任务 ID 示例：8b5425ec-9f2a-43bd-a3ab-01912f50282b；任务链接 URL 示例：https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    llm_description: 要删除的任务的 GUID，支持传任务 ID 和任务链接 URL。任务 ID 示例：8b5425ec-9f2a-43bd-a3ab-01912f50282b；任务链接 URL 示例：https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_task/tools/update_task.py b/api/core/tools/provider/builtin/lark_task/tools/update_task.py
new file mode 100644
index 0000000000..0d3469c91a
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/update_task.py
@@ -0,0 +1,23 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class UpdateTaskTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        task_guid = tool_parameters.get("task_guid")
+        summary = tool_parameters.get("summary")
+        start_time = tool_parameters.get("start_time")
+        end_time = tool_parameters.get("end_time")
+        completed_time = tool_parameters.get("completed_time")
+        description = tool_parameters.get("description")
+
+        res = client.update_task(task_guid, summary, start_time, end_time, completed_time, description)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_task/tools/update_task.yaml b/api/core/tools/provider/builtin/lark_task/tools/update_task.yaml
new file mode 100644
index 0000000000..a98f037f21
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_task/tools/update_task.yaml
@@ -0,0 +1,89 @@
+identity:
+  name: update_task
+  author: Doug Lea
+  label:
+    en_US: Update Lark Task
+    zh_Hans: 更新 Lark 任务
+description:
+  human:
+    en_US: Update Lark Task
+    zh_Hans: 更新 Lark 任务
+  llm: A tool for updating tasks in Lark.(更新 Lark 任务)
+parameters:
+  - name: task_guid
+    type: string
+    required: true
+    label:
+      en_US: Task GUID
+      zh_Hans: 任务 GUID
+    human_description:
+      en_US: |
+        The task ID, supports inputting either the Task ID or the Task link URL. Example of Task ID: 42cad8a0-f8c8-4344-9be2-d1d7e8e91b64; Example of Task link URL: https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+      zh_Hans: |
+        任务ID，支持传入任务 ID 和任务链接 URL。任务 ID 示例: 42cad8a0-f8c8-4344-9be2-d1d7e8e91b64；任务链接 URL 示例: https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    llm_description: |
+      任务ID，支持传入任务 ID 和任务链接 URL。任务 ID 示例: 42cad8a0-f8c8-4344-9be2-d1d7e8e91b64；任务链接 URL 示例: https://applink.larksuite.com/client/todo/detail?guid=1b066afa-96de-406c-90a3-dfd30159a571&suite_entity_num=t100805
+    form: llm
+
+  - name: summary
+    type: string
+    required: true
+    label:
+      en_US: Task Title
+      zh_Hans: 任务标题
+    human_description:
+      en_US: The title of the task.
+      zh_Hans: 任务标题
+    llm_description: 任务标题
+    form: llm
+
+  - name: description
+    type: string
+    required: false
+    label:
+      en_US: Task Description
+      zh_Hans: 任务备注
+    human_description:
+      en_US: The description or notes for the task.
+      zh_Hans: 任务备注
+    llm_description: 任务备注
+    form: llm
+
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: Start Time
+      zh_Hans: 任务开始时间
+    human_description:
+      en_US: |
+        The start time of the task, in the format: 2006-01-02 15:04:05
+      zh_Hans: 任务开始时间，格式为：2006-01-02 15:04:05
+    llm_description: 任务开始时间，格式为：2006-01-02 15:04:05
+    form: llm
+
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: End Time
+      zh_Hans: 任务结束时间
+    human_description:
+      en_US: |
+        The end time of the task, in the format: 2006-01-02 15:04:05
+      zh_Hans: 任务结束时间，格式为：2006-01-02 15:04:05
+    llm_description: 任务结束时间，格式为：2006-01-02 15:04:05
+    form: llm
+
+  - name: completed_time
+    type: string
+    required: false
+    label:
+      en_US: Completed Time
+      zh_Hans: 任务完成时间
+    human_description:
+      en_US: |
+        The completion time of the task, in the format: 2006-01-02 15:04:05
+      zh_Hans: 任务完成时间，格式为：2006-01-02 15:04:05
+    llm_description: 任务完成时间，格式为：2006-01-02 15:04:05
+    form: llm
diff --git a/api/core/tools/provider/builtin/lark_wiki/_assets/icon.png b/api/core/tools/provider/builtin/lark_wiki/_assets/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..47f6b8c30ea0cf2d4b625bf5890ab94b0b5ca84e
GIT binary patch
literal 2827
zcmV+m3-t7fP)<h;3K|Lk000e1NJLTq002M$002M;1^@s6s%dfF000WoNkl<ZcwX&V
z+jCac755)#r|;##SA8j++Cr<9pd#K0B2X_tL_tuINI*gI-NLO<D&ZnmB@k{=3AbQG
zaX8aXpX*!2TO+==JamR6AFx-~T6?W?*4gJICFC%*o$Q(4tozz)t+UTQA53n$ZETIL
zu{E~F*6^#j1AemN2#jBT2pR}018R|{yBU25^`kndp0LJSwpMn)?JY;3Z}|}bi<YC{
zEKNam#%BF6Up8hK#QrA7ilgv3>f5Qlu;xM+Zqm_m7+@I|v5ctjtUH9oc^%8b$52^q
znkS?h&%EfVX&WE=g!xF1H8$U>L9`r#=yhxq;izHZ8KG=+-WQ8weW%y4{&3){J{I0T
z0r}+l8DQSNTQY>Dop7NU3$maBna73<^}MxvSf4~5Inr@<?<DtS9K!gphf%Gi2OxUw
zAh@Lm0bV-*Ky;~654Md6ok@)54*5_fK2=9&B5OXDJK>*;iaZHRXaEQNrc0D|Br@|<
zoY?Zfep#KfI+DYEGC1gCI96(4f8>{#8mQ*5kDFeYJbD#3An-=Qzludzy~tiC1;+Az
z<7ab--N?xvCclOC&BV(-gje>%ys(#>TCd;$Fc~0qOasuS8Tma=U*4aMiFHXnN4h*;
zyC>%zqJ|uhUooeE61|KK<zexA;1<6t{JUOe8#N?!F@f|-Z4T1eILV^}$=ff~r*iI3
z>lW-S;6lWLA}$iKh#*ZP?+CFzZBzf0UGtcHj$$O2_uH4Uhcxk%I9m%t;+gJ)=CGd;
zFH97@v=1Bt0`O)p;iimGCw?rWexcgZ`k30hDY`dfOKLIuus-cegXswS8Sz@U2j*e%
z6{OV^HT;WvM9$j1oD;(~)u~==*ho7z5G~w`gTtm$pkW$*LE=L+<&Sxaxd;ARYtb&a
zeZg+%Sg;4*=OKW@ccCuU|E~`~7oD&db_2YCO}$8Tmk06GDBFlvTjKlwF#uZFSo{2)
z0P}Y-oKRrhZX+jf7&H0-G`#~&h>D=vv@7N<s`JT~pwhJ6^Q;%1DaHF7p<H=>C(LWw
zGW5?_wD&db1oylL5HOCZAl3^C7{CsQa1$rs4d)p?#twn<R<CiwQ+%8w^d;<sEjWnG
ze|n(LQt@}G{jhODCT9@?z@j56eC`f#1mYR*B=X$tq7z;b+mtu)n$N=Kn6o`wJe>{`
zpXAX!gB(aRIYzvw5QLlKiz4dgY!lv758_p?I*H+i#3s#oAC1`-H|_IFZfq#?dp?yX
z*)ORn&5wQn&uxPU;hx(HfK=h-5Ty6;exj#>h(4>f#5atI^NMe#0nU}=NY|6dicI{7
z*R$;qJ+l=Y0eK)WztvOat&Nf6N4nV8<P$mek8>e%>hXTtB)8_2IIWXtAp>}N3q()1
zgF_0>Ac!*GK0MF96o-i8B#-5-pUt6pa_ULPtFOi@@8tpp@HSld-P-^NN1#oVk3l+t
z*cp}=-;~YR)uuUNHc@OMtS8Qq#3y;v`;k2Q0bn7~>_1{b1S%}cyO@`aXQ@v4FeZo<
zpTtjWq|el&@sdMh%|4!bpV1Fs=4OZxK$x`|98xz+WaVd3A>Qu&d|@#!v1+Gc<JdTd
z&#nHkEvFu}YaW&rx)0fAZ!T~EPyHdmOhd)yST=q$y^fo*J_6-^obXDmZ&;7X5#~(A
zO0LllU<M|cu?axn1T51=PX;=I=wcajBz$Na9+?x6lP@o?@}nQX%ylsSsSVK3h`@#$
zh`$j`-vpD0&e#ZzIFODfH{xK@a5unbP5|3C4Vfo}LIzOOYTN|lr~e)<BcRd45lB{q
zI`YfpF}`u^#&oRl`QZ$J7M`LeZhnKuv<={<2^BfRhdD^Ev5h{S{aLr(=Zkay&;~#Y
zi@f{<7K#9(RIriZiFJlwpFIJO2X=!vC(&ty4FwEfZP)dcYmQw%VA0lf?I5AnRqw2Q
zfH;1wvwEU_h)_9IKg{SE-1tWKm2qLd$JaxIO#lOnd~7YaST=ZUo!4jLb9~uf@}%;u
zFJJ)N@UiQEv#1@r`bFFO^@}yahlu0#uPWX3uWBcdo~(aeJvI1E_4MF3l`{i>+V~gv
zk<Et%Mvr2V2r3qsU&C@rhXVl8+gOSBcJZ}PJ>`*Hw_pHWIDjr3zz6k<ZP<MEc>T+o
zG+#YAh|S{wu=#Qi4xne?+sdiIJ8eE}DtcrM0KyS{)bNkA@i16cYGY|hti~n4SQ{@s
zMGxTW7t}mGd1}6P{OUz+o^TQeAk9~$`EpPH-%B6WC)j)sw?Tw(2za;_P!G#zUIEo(
zv+&S=jxTlq+I)?kJ~m(Nu762KzuGNN9*=&Rp8gqp@@H`XX9qF^n2JSV;Z*#Pn2HTO
z<QvQkfbCVZt1t5G@A-7DA_t)Fy+8U_(??&L*P~yir_asz_mw`$4&Xs7lt2Y59;8Nu
zDo?Pn93S}F97d=5f(C%^JvT2;pN_s>dR4si@YesjOi!PV{u$qV8JjQn4rB%}1&f+e
z0hr<&^;Bfm3)oi44j}YnU-d7GP3Y800Rw0|N}oIP^y$4n?%#X8^r(4y@5^TiIDn6E
z0KGVXj|ckNf&n~$MLkdk5V%~v#Js~sG6Tr<&&GOPAp@ZAy=`7Dy-INO`N{Lqr;E=&
z{qnhizDn<vi8db=7~Nk2xUUs}DDzk}%I^Mz#x@&EOYQebob?T35d{q3ef-fQ?|m)v
z$xGgQzVv$Z==-OS1Nd(B`SbuLw?agPPG0SYK)lcxlRa2(V<n#LvF~c{r~W~X(GQ^7
zHMqfl^861TZJytI`Q-7F=T9D9eA0aB<ALu=IDkqpfQHo&A$R}{gtx$@r~&!dSdI_<
zysS{Y<{$k4+CIECSKoWS`1E_PN8jK0$$L*19}NJTFMry9DLes_RzY;{DlGObfJsAu
z?8@6%j#s}`#us5@5Tq++0IgmBx}(;8HTRRpKX~MOAK&`YeCd<^OQoR)K!tPhzg_7;
zvhZqS*jQS|&)pmNG9(xKttwyuwCEo>xt~0_^z5xq&9~A3u=&#Y{!8I2aL-DJs7b;-
zLYaRP9LfZhS!ZJ<J{u>#CXdo$22eY7ZF)t2=gTKgkA8`sK0SGV>-*;E>7OTD`93@W
zcfSb{6?`}04RB1AH&7oNEAc!WiSab8(;TELW&pJ4pSXFSJagmo0QBhd06x3&eI=@=
z2e1O7-(YbuR)Aw9e9C5HX~lM=zZPC8!sn8Fv{x8^(`#)H{Iqg<@UP)}Pq%&r2S880
zEX|j(`Br@TgzyAlu@Mzan7ACA@)>zK%b63Peq;;dLP~pu@uOO;CkF2*pC0&JkG?j~
zpFH#AwRu7%N)KSd3b;VUPiXNC8s1GHo^&n9xA-P)EG>;m#U=jAt52~5pjGL<G7c}j
zKht|p&6m#&_LV>NmmWQNdhdrFKnp~_!U7q~1j>8rVtY$=0BrYhM7h~GHl2qCRQv#F
z;m_M~<=(-P-um?P>6;%u{k#F(^*Thq#Nq@FWyidyiR^Wm0T`PjJ5i4Dyyo?2F5vG3
zhg&p=(z*UJo`7C`@52)i-T^d#yP5%zDyRV~1KSb5XwF`OsON@bia@{X%8N|&+lkFX
z|GiP+tJ3?AAFujs@J_n*scHKDGey1i`}s2X85ZAgqp-2GLVF%)kMYgfzcCtHV{2@U
dt+5rXe*^9v#%`^xa$f)d002ovPDHLkV1jYtfBFCb

literal 0
HcmV?d00001

diff --git a/api/core/tools/provider/builtin/lark_wiki/lark_wiki.py b/api/core/tools/provider/builtin/lark_wiki/lark_wiki.py
new file mode 100644
index 0000000000..e6941206ee
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_wiki/lark_wiki.py
@@ -0,0 +1,7 @@
+from core.tools.provider.builtin_tool_provider import BuiltinToolProviderController
+from core.tools.utils.lark_api_utils import lark_auth
+
+
+class LarkWikiProvider(BuiltinToolProviderController):
+    def _validate_credentials(self, credentials: dict) -> None:
+        lark_auth(credentials)
diff --git a/api/core/tools/provider/builtin/lark_wiki/lark_wiki.yaml b/api/core/tools/provider/builtin/lark_wiki/lark_wiki.yaml
new file mode 100644
index 0000000000..86bef00086
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_wiki/lark_wiki.yaml
@@ -0,0 +1,36 @@
+identity:
+  author: Doug Lea
+  name: lark_wiki
+  label:
+    en_US: Lark Wiki
+    zh_Hans: Lark 知识库
+  description:
+    en_US: |
+      Lark Wiki, requires the following permissions: wiki:wiki:readonly.
+    zh_Hans: |
+      Lark 知识库，需要开通以下权限: wiki:wiki:readonly。
+  icon: icon.png
+  tags:
+    - social
+    - productivity
+credentials_for_provider:
+  app_id:
+    type: text-input
+    required: true
+    label:
+      en_US: APP ID
+    placeholder:
+      en_US: Please input your Lark app id
+      zh_Hans: 请输入你的 Lark app id
+    help:
+      en_US: Get your app_id and app_secret from Lark
+      zh_Hans: 从 Lark 获取您的 app_id 和 app_secret
+    url: https://open.larksuite.com/app
+  app_secret:
+    type: secret-input
+    required: true
+    label:
+      en_US: APP Secret
+    placeholder:
+      en_US: Please input your app secret
+      zh_Hans: 请输入你的 Lark app secret
diff --git a/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.py b/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.py
new file mode 100644
index 0000000000..a05f300755
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.py
@@ -0,0 +1,21 @@
+from typing import Any
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+from core.tools.utils.lark_api_utils import LarkRequest
+
+
+class GetWikiNodesTool(BuiltinTool):
+    def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> ToolInvokeMessage:
+        app_id = self.runtime.credentials.get("app_id")
+        app_secret = self.runtime.credentials.get("app_secret")
+        client = LarkRequest(app_id, app_secret)
+
+        space_id = tool_parameters.get("space_id")
+        parent_node_token = tool_parameters.get("parent_node_token")
+        page_token = tool_parameters.get("page_token")
+        page_size = tool_parameters.get("page_size")
+
+        res = client.get_wiki_nodes(space_id, parent_node_token, page_token, page_size)
+
+        return self.create_json_message(res)
diff --git a/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.yaml b/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.yaml
new file mode 100644
index 0000000000..a8c242a2e9
--- /dev/null
+++ b/api/core/tools/provider/builtin/lark_wiki/tools/get_wiki_nodes.yaml
@@ -0,0 +1,63 @@
+identity:
+  name: get_wiki_nodes
+  author: Doug Lea
+  label:
+    en_US: Get Wiki Nodes
+    zh_Hans: 获取知识空间子节点列表
+description:
+  human:
+    en_US: |
+      Get the list of child nodes in Wiki, make sure the app/bot is a member of the wiki space. See How to add an app as a wiki base administrator (member). https://open.larksuite.com/document/server-docs/docs/wiki-v2/wiki-qa
+    zh_Hans: |
+      获取知识库全部子节点列表，请确保应用/机器人为知识空间成员。参阅如何将应用添加为知识库管理员（成员）。https://open.larksuite.com/document/server-docs/docs/wiki-v2/wiki-qa
+  llm: A tool for getting all sub-nodes of a knowledge base.(获取知识空间子节点列表)
+parameters:
+  - name: space_id
+    type: string
+    required: true
+    label:
+      en_US: Space Id
+      zh_Hans: 知识空间 ID
+    human_description:
+      en_US: |
+        The ID of the knowledge space. Supports space link URL, for example: https://lark-japan.jp.larksuite.com/wiki/settings/7431084851517718561
+      zh_Hans: 知识空间 ID，支持空间链接 URL，例如：https://lark-japan.jp.larksuite.com/wiki/settings/7431084851517718561
+    llm_description: 知识空间 ID，支持空间链接 URL，例如：https://lark-japan.jp.larksuite.com/wiki/settings/7431084851517718561
+    form: llm
+
+  - name: page_size
+    type: number
+    required: false
+    default: 10
+    label:
+      en_US: Page Size
+      zh_Hans: 分页大小
+    human_description:
+      en_US: The size of each page, with a maximum value of 50.
+      zh_Hans: 分页大小，最大值 50。
+    llm_description: 分页大小，最大值 50。
+    form: form
+
+  - name: page_token
+    type: string
+    required: false
+    label:
+      en_US: Page Token
+      zh_Hans: 分页标记
+    human_description:
+      en_US: The pagination token. Leave empty for the first request to start from the beginning; if the paginated query result has more items, a new page_token will be returned, which can be used to get the next set of results.
+      zh_Hans: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    llm_description: 分页标记，第一次请求不填，表示从头开始遍历；分页查询结果还有更多项时会同时返回新的 page_token，下次遍历可采用该 page_token 获取查询结果。
+    form: llm
+
+  - name: parent_node_token
+    type: string
+    required: false
+    label:
+      en_US: Parent Node Token
+      zh_Hans: 父节点 token
+    human_description:
+      en_US: The token of the parent node.
+      zh_Hans: 父节点 token
+    llm_description: 父节点 token
+    form: llm
diff --git a/api/core/tools/utils/feishu_api_utils.py b/api/core/tools/utils/feishu_api_utils.py
index 722cf4b538..ea28037df0 100644
--- a/api/core/tools/utils/feishu_api_utils.py
+++ b/api/core/tools/utils/feishu_api_utils.py
@@ -127,7 +127,9 @@ class FeishuRequest:
             "folder_token": folder_token,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def write_document(self, document_id: str, content: str, position: str = "end") -> dict:
         url = f"{self.API_BASE_URL}/document/write_document"
@@ -135,7 +137,7 @@ class FeishuRequest:
         res = self._send_request(url, payload=payload)
         return res
 
-    def get_document_content(self, document_id: str, mode: str = "markdown", lang: str = "0") -> dict:
+    def get_document_content(self, document_id: str, mode: str = "markdown", lang: str = "0") -> str:
         """
         API url: https://open.larkoffice.com/document/server-docs/docs/docs/docx-v1/document/raw_content
         Example Response:
@@ -154,7 +156,9 @@ class FeishuRequest:
         }
         url = f"{self.API_BASE_URL}/document/get_document_content"
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data").get("content")
+        if "data" in res:
+            return res.get("data").get("content")
+        return ""
 
     def list_document_blocks(
         self, document_id: str, page_token: str, user_id_type: str = "open_id", page_size: int = 500
@@ -170,7 +174,9 @@ class FeishuRequest:
         }
         url = f"{self.API_BASE_URL}/document/list_document_blocks"
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def send_bot_message(self, receive_id_type: str, receive_id: str, msg_type: str, content: str) -> dict:
         """
@@ -186,7 +192,9 @@ class FeishuRequest:
             "content": content.strip('"').replace(r"\"", '"').replace(r"\\", "\\"),
         }
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def send_webhook_message(self, webhook: str, msg_type: str, content: str) -> dict:
         url = f"{self.API_BASE_URL}/message/send_webhook_message"
@@ -220,7 +228,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def get_thread_messages(
         self, container_id: str, page_token: str, sort_type: str = "ByCreateTimeAsc", page_size: int = 20
@@ -236,7 +246,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def create_task(self, summary: str, start_time: str, end_time: str, completed_time: str, description: str) -> dict:
         # 创建任务
@@ -249,7 +261,9 @@ class FeishuRequest:
             "description": description,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def update_task(
         self, task_guid: str, summary: str, start_time: str, end_time: str, completed_time: str, description: str
@@ -265,7 +279,9 @@ class FeishuRequest:
             "description": description,
         }
         res = self._send_request(url, method="PATCH", payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def delete_task(self, task_guid: str) -> dict:
         # 删除任务
@@ -297,7 +313,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def get_primary_calendar(self, user_id_type: str = "open_id") -> dict:
         url = f"{self.API_BASE_URL}/calendar/get_primary_calendar"
@@ -305,7 +323,9 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def create_event(
         self,
@@ -328,7 +348,9 @@ class FeishuRequest:
             "attendee_ability": attendee_ability,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def update_event(
         self,
@@ -374,7 +396,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def search_events(
         self,
@@ -395,7 +419,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def add_event_attendees(self, event_id: str, attendee_phone_or_email: str, need_notification: bool = True) -> dict:
         # 参加日程参会人
@@ -406,7 +432,9 @@ class FeishuRequest:
             "need_notification": need_notification,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def create_spreadsheet(
         self,
@@ -420,7 +448,9 @@ class FeishuRequest:
             "folder_token": folder_token,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def get_spreadsheet(
         self,
@@ -434,7 +464,9 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def list_spreadsheet_sheets(
         self,
@@ -446,7 +478,9 @@ class FeishuRequest:
             "spreadsheet_token": spreadsheet_token,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def add_rows(
         self,
@@ -466,7 +500,9 @@ class FeishuRequest:
             "values": values,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def add_cols(
         self,
@@ -486,7 +522,9 @@ class FeishuRequest:
             "values": values,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def read_rows(
         self,
@@ -508,7 +546,9 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def read_cols(
         self,
@@ -530,7 +570,9 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def read_table(
         self,
@@ -552,7 +594,9 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def create_base(
         self,
@@ -566,7 +610,9 @@ class FeishuRequest:
             "folder_token": folder_token,
         }
         res = self._send_request(url, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def add_records(
         self,
@@ -588,7 +634,9 @@ class FeishuRequest:
             "records": convert_add_records(records),
         }
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def update_records(
         self,
@@ -610,7 +658,9 @@ class FeishuRequest:
             "records": convert_update_records(records),
         }
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def delete_records(
         self,
@@ -637,7 +687,9 @@ class FeishuRequest:
             "records": record_id_list,
         }
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def search_record(
         self,
@@ -701,7 +753,10 @@ class FeishuRequest:
         if automatic_fields:
             payload["automatic_fields"] = automatic_fields
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def get_base_info(
         self,
@@ -713,7 +768,9 @@ class FeishuRequest:
             "app_token": app_token,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def create_table(
         self,
@@ -741,7 +798,9 @@ class FeishuRequest:
         if default_view_name:
             payload["default_view_name"] = default_view_name
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def delete_tables(
         self,
@@ -774,8 +833,11 @@ class FeishuRequest:
             "table_ids": table_id_list,
             "table_names": table_name_list,
         }
+
         res = self._send_request(url, params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def list_tables(
         self,
@@ -791,7 +853,9 @@ class FeishuRequest:
             "page_size": page_size,
         }
         res = self._send_request(url, method="GET", params=params)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
 
     def read_records(
         self,
@@ -819,4 +883,6 @@ class FeishuRequest:
             "user_id_type": user_id_type,
         }
         res = self._send_request(url, method="GET", params=params, payload=payload)
-        return res.get("data")
+        if "data" in res:
+            return res.get("data")
+        return res
diff --git a/api/core/tools/utils/lark_api_utils.py b/api/core/tools/utils/lark_api_utils.py
new file mode 100644
index 0000000000..30cb0cb141
--- /dev/null
+++ b/api/core/tools/utils/lark_api_utils.py
@@ -0,0 +1,820 @@
+import json
+from typing import Optional
+
+import httpx
+
+from core.tools.errors import ToolProviderCredentialValidationError
+from extensions.ext_redis import redis_client
+
+
+def lark_auth(credentials):
+    app_id = credentials.get("app_id")
+    app_secret = credentials.get("app_secret")
+    if not app_id or not app_secret:
+        raise ToolProviderCredentialValidationError("app_id and app_secret is required")
+    try:
+        assert LarkRequest(app_id, app_secret).tenant_access_token is not None
+    except Exception as e:
+        raise ToolProviderCredentialValidationError(str(e))
+
+
+class LarkRequest:
+    API_BASE_URL = "https://lark-plugin-api.solutionsuite.ai/lark-plugin"
+
+    def __init__(self, app_id: str, app_secret: str):
+        self.app_id = app_id
+        self.app_secret = app_secret
+
+    def convert_add_records(self, json_str):
+        try:
+            data = json.loads(json_str)
+            if not isinstance(data, list):
+                raise ValueError("Parsed data must be a list")
+            converted_data = [{"fields": json.dumps(item, ensure_ascii=False)} for item in data]
+            return converted_data
+        except json.JSONDecodeError:
+            raise ValueError("The input string is not valid JSON")
+        except Exception as e:
+            raise ValueError(f"An error occurred while processing the data: {e}")
+
+    def convert_update_records(self, json_str):
+        try:
+            data = json.loads(json_str)
+            if not isinstance(data, list):
+                raise ValueError("Parsed data must be a list")
+
+            converted_data = [
+                {"fields": json.dumps(record["fields"], ensure_ascii=False), "record_id": record["record_id"]}
+                for record in data
+                if "fields" in record and "record_id" in record
+            ]
+
+            if len(converted_data) != len(data):
+                raise ValueError("Each record must contain 'fields' and 'record_id'")
+
+            return converted_data
+        except json.JSONDecodeError:
+            raise ValueError("The input string is not valid JSON")
+        except Exception as e:
+            raise ValueError(f"An error occurred while processing the data: {e}")
+
+    @property
+    def tenant_access_token(self) -> str:
+        feishu_tenant_access_token = f"tools:{self.app_id}:feishu_tenant_access_token"
+        if redis_client.exists(feishu_tenant_access_token):
+            return redis_client.get(feishu_tenant_access_token).decode()
+        res = self.get_tenant_access_token(self.app_id, self.app_secret)
+        redis_client.setex(feishu_tenant_access_token, res.get("expire"), res.get("tenant_access_token"))
+        if "tenant_access_token" in res:
+            return res.get("tenant_access_token")
+        return ""
+
+    def _send_request(
+        self,
+        url: str,
+        method: str = "post",
+        require_token: bool = True,
+        payload: Optional[dict] = None,
+        params: Optional[dict] = None,
+    ):
+        headers = {
+            "Content-Type": "application/json",
+            "user-agent": "Dify",
+        }
+        if require_token:
+            headers["tenant-access-token"] = f"{self.tenant_access_token}"
+        res = httpx.request(method=method, url=url, headers=headers, json=payload, params=params, timeout=30).json()
+        if res.get("code") != 0:
+            raise Exception(res)
+        return res
+
+    def get_tenant_access_token(self, app_id: str, app_secret: str) -> dict:
+        url = f"{self.API_BASE_URL}/access_token/get_tenant_access_token"
+        payload = {"app_id": app_id, "app_secret": app_secret}
+        res = self._send_request(url, require_token=False, payload=payload)
+        return res
+
+    def create_document(self, title: str, content: str, folder_token: str) -> dict:
+        url = f"{self.API_BASE_URL}/document/create_document"
+        payload = {
+            "title": title,
+            "content": content,
+            "folder_token": folder_token,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def write_document(self, document_id: str, content: str, position: str = "end") -> dict:
+        url = f"{self.API_BASE_URL}/document/write_document"
+        payload = {"document_id": document_id, "content": content, "position": position}
+        res = self._send_request(url, payload=payload)
+        return res
+
+    def get_document_content(self, document_id: str, mode: str = "markdown", lang: str = "0") -> str | dict:
+        params = {
+            "document_id": document_id,
+            "mode": mode,
+            "lang": lang,
+        }
+        url = f"{self.API_BASE_URL}/document/get_document_content"
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data").get("content")
+        return ""
+
+    def list_document_blocks(
+        self, document_id: str, page_token: str, user_id_type: str = "open_id", page_size: int = 500
+    ) -> dict:
+        params = {
+            "user_id_type": user_id_type,
+            "document_id": document_id,
+            "page_size": page_size,
+            "page_token": page_token,
+        }
+        url = f"{self.API_BASE_URL}/document/list_document_blocks"
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def send_bot_message(self, receive_id_type: str, receive_id: str, msg_type: str, content: str) -> dict:
+        url = f"{self.API_BASE_URL}/message/send_bot_message"
+        params = {
+            "receive_id_type": receive_id_type,
+        }
+        payload = {
+            "receive_id": receive_id,
+            "msg_type": msg_type,
+            "content": content.strip('"').replace(r"\"", '"').replace(r"\\", "\\"),
+        }
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def send_webhook_message(self, webhook: str, msg_type: str, content: str) -> dict:
+        url = f"{self.API_BASE_URL}/message/send_webhook_message"
+        payload = {
+            "webhook": webhook,
+            "msg_type": msg_type,
+            "content": content.strip('"').replace(r"\"", '"').replace(r"\\", "\\"),
+        }
+        res = self._send_request(url, require_token=False, payload=payload)
+        return res
+
+    def get_chat_messages(
+        self,
+        container_id: str,
+        start_time: str,
+        end_time: str,
+        page_token: str,
+        sort_type: str = "ByCreateTimeAsc",
+        page_size: int = 20,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/message/get_chat_messages"
+        params = {
+            "container_id": container_id,
+            "start_time": start_time,
+            "end_time": end_time,
+            "sort_type": sort_type,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def get_thread_messages(
+        self, container_id: str, page_token: str, sort_type: str = "ByCreateTimeAsc", page_size: int = 20
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/message/get_thread_messages"
+        params = {
+            "container_id": container_id,
+            "sort_type": sort_type,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def create_task(self, summary: str, start_time: str, end_time: str, completed_time: str, description: str) -> dict:
+        url = f"{self.API_BASE_URL}/task/create_task"
+        payload = {
+            "summary": summary,
+            "start_time": start_time,
+            "end_time": end_time,
+            "completed_at": completed_time,
+            "description": description,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def update_task(
+        self, task_guid: str, summary: str, start_time: str, end_time: str, completed_time: str, description: str
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/task/update_task"
+        payload = {
+            "task_guid": task_guid,
+            "summary": summary,
+            "start_time": start_time,
+            "end_time": end_time,
+            "completed_time": completed_time,
+            "description": description,
+        }
+        res = self._send_request(url, method="PATCH", payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def delete_task(self, task_guid: str) -> dict:
+        url = f"{self.API_BASE_URL}/task/delete_task"
+        payload = {
+            "task_guid": task_guid,
+        }
+        res = self._send_request(url, method="DELETE", payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def add_members(self, task_guid: str, member_phone_or_email: str, member_role: str) -> dict:
+        url = f"{self.API_BASE_URL}/task/add_members"
+        payload = {
+            "task_guid": task_guid,
+            "member_phone_or_email": member_phone_or_email,
+            "member_role": member_role,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def get_wiki_nodes(self, space_id: str, parent_node_token: str, page_token: str, page_size: int = 20) -> dict:
+        url = f"{self.API_BASE_URL}/wiki/get_wiki_nodes"
+        payload = {
+            "space_id": space_id,
+            "parent_node_token": parent_node_token,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def get_primary_calendar(self, user_id_type: str = "open_id") -> dict:
+        url = f"{self.API_BASE_URL}/calendar/get_primary_calendar"
+        params = {
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def create_event(
+        self,
+        summary: str,
+        description: str,
+        start_time: str,
+        end_time: str,
+        attendee_ability: str,
+        need_notification: bool = True,
+        auto_record: bool = False,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/create_event"
+        payload = {
+            "summary": summary,
+            "description": description,
+            "need_notification": need_notification,
+            "start_time": start_time,
+            "end_time": end_time,
+            "auto_record": auto_record,
+            "attendee_ability": attendee_ability,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def update_event(
+        self,
+        event_id: str,
+        summary: str,
+        description: str,
+        need_notification: bool,
+        start_time: str,
+        end_time: str,
+        auto_record: bool,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/update_event/{event_id}"
+        payload = {}
+        if summary:
+            payload["summary"] = summary
+        if description:
+            payload["description"] = description
+        if start_time:
+            payload["start_time"] = start_time
+        if end_time:
+            payload["end_time"] = end_time
+        if need_notification:
+            payload["need_notification"] = need_notification
+        if auto_record:
+            payload["auto_record"] = auto_record
+        res = self._send_request(url, method="PATCH", payload=payload)
+        return res
+
+    def delete_event(self, event_id: str, need_notification: bool = True) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/delete_event/{event_id}"
+        params = {
+            "need_notification": need_notification,
+        }
+        res = self._send_request(url, method="DELETE", params=params)
+        return res
+
+    def list_events(self, start_time: str, end_time: str, page_token: str, page_size: int = 50) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/list_events"
+        params = {
+            "start_time": start_time,
+            "end_time": end_time,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def search_events(
+        self,
+        query: str,
+        start_time: str,
+        end_time: str,
+        page_token: str,
+        user_id_type: str = "open_id",
+        page_size: int = 20,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/search_events"
+        payload = {
+            "query": query,
+            "start_time": start_time,
+            "end_time": end_time,
+            "page_token": page_token,
+            "user_id_type": user_id_type,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def add_event_attendees(self, event_id: str, attendee_phone_or_email: str, need_notification: bool = True) -> dict:
+        url = f"{self.API_BASE_URL}/calendar/add_event_attendees"
+        payload = {
+            "event_id": event_id,
+            "attendee_phone_or_email": attendee_phone_or_email,
+            "need_notification": need_notification,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def create_spreadsheet(
+        self,
+        title: str,
+        folder_token: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/create_spreadsheet"
+        payload = {
+            "title": title,
+            "folder_token": folder_token,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def get_spreadsheet(
+        self,
+        spreadsheet_token: str,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/get_spreadsheet"
+        params = {
+            "spreadsheet_token": spreadsheet_token,
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def list_spreadsheet_sheets(
+        self,
+        spreadsheet_token: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/list_spreadsheet_sheets"
+        params = {
+            "spreadsheet_token": spreadsheet_token,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def add_rows(
+        self,
+        spreadsheet_token: str,
+        sheet_id: str,
+        sheet_name: str,
+        length: int,
+        values: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/add_rows"
+        payload = {
+            "spreadsheet_token": spreadsheet_token,
+            "sheet_id": sheet_id,
+            "sheet_name": sheet_name,
+            "length": length,
+            "values": values,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def add_cols(
+        self,
+        spreadsheet_token: str,
+        sheet_id: str,
+        sheet_name: str,
+        length: int,
+        values: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/add_cols"
+        payload = {
+            "spreadsheet_token": spreadsheet_token,
+            "sheet_id": sheet_id,
+            "sheet_name": sheet_name,
+            "length": length,
+            "values": values,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def read_rows(
+        self,
+        spreadsheet_token: str,
+        sheet_id: str,
+        sheet_name: str,
+        start_row: int,
+        num_rows: int,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/read_rows"
+        params = {
+            "spreadsheet_token": spreadsheet_token,
+            "sheet_id": sheet_id,
+            "sheet_name": sheet_name,
+            "start_row": start_row,
+            "num_rows": num_rows,
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def read_cols(
+        self,
+        spreadsheet_token: str,
+        sheet_id: str,
+        sheet_name: str,
+        start_col: int,
+        num_cols: int,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/read_cols"
+        params = {
+            "spreadsheet_token": spreadsheet_token,
+            "sheet_id": sheet_id,
+            "sheet_name": sheet_name,
+            "start_col": start_col,
+            "num_cols": num_cols,
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def read_table(
+        self,
+        spreadsheet_token: str,
+        sheet_id: str,
+        sheet_name: str,
+        num_range: str,
+        query: str,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/spreadsheet/read_table"
+        params = {
+            "spreadsheet_token": spreadsheet_token,
+            "sheet_id": sheet_id,
+            "sheet_name": sheet_name,
+            "range": num_range,
+            "query": query,
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def create_base(
+        self,
+        name: str,
+        folder_token: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/create_base"
+        payload = {
+            "name": name,
+            "folder_token": folder_token,
+        }
+        res = self._send_request(url, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def add_records(
+        self,
+        app_token: str,
+        table_id: str,
+        table_name: str,
+        records: str,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/add_records"
+        params = {
+            "app_token": app_token,
+            "table_id": table_id,
+            "table_name": table_name,
+            "user_id_type": user_id_type,
+        }
+        payload = {
+            "records": self.convert_add_records(records),
+        }
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def update_records(
+        self,
+        app_token: str,
+        table_id: str,
+        table_name: str,
+        records: str,
+        user_id_type: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/update_records"
+        params = {
+            "app_token": app_token,
+            "table_id": table_id,
+            "table_name": table_name,
+            "user_id_type": user_id_type,
+        }
+        payload = {
+            "records": self.convert_update_records(records),
+        }
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def delete_records(
+        self,
+        app_token: str,
+        table_id: str,
+        table_name: str,
+        record_ids: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/delete_records"
+        params = {
+            "app_token": app_token,
+            "table_id": table_id,
+            "table_name": table_name,
+        }
+        if not record_ids:
+            record_id_list = []
+        else:
+            try:
+                record_id_list = json.loads(record_ids)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+        payload = {
+            "records": record_id_list,
+        }
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def search_record(
+        self,
+        app_token: str,
+        table_id: str,
+        table_name: str,
+        view_id: str,
+        field_names: str,
+        sort: str,
+        filters: str,
+        page_token: str,
+        automatic_fields: bool = False,
+        user_id_type: str = "open_id",
+        page_size: int = 20,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/search_record"
+
+        params = {
+            "app_token": app_token,
+            "table_id": table_id,
+            "table_name": table_name,
+            "user_id_type": user_id_type,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+
+        if not field_names:
+            field_name_list = []
+        else:
+            try:
+                field_name_list = json.loads(field_names)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+
+        if not sort:
+            sort_list = []
+        else:
+            try:
+                sort_list = json.loads(sort)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+
+        if not filters:
+            filter_dict = {}
+        else:
+            try:
+                filter_dict = json.loads(filters)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+
+        payload = {}
+
+        if view_id:
+            payload["view_id"] = view_id
+        if field_names:
+            payload["field_names"] = field_name_list
+        if sort:
+            payload["sort"] = sort_list
+        if filters:
+            payload["filter"] = filter_dict
+        if automatic_fields:
+            payload["automatic_fields"] = automatic_fields
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def get_base_info(
+        self,
+        app_token: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/get_base_info"
+        params = {
+            "app_token": app_token,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def create_table(
+        self,
+        app_token: str,
+        table_name: str,
+        default_view_name: str,
+        fields: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/create_table"
+        params = {
+            "app_token": app_token,
+        }
+        if not fields:
+            fields_list = []
+        else:
+            try:
+                fields_list = json.loads(fields)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+        payload = {
+            "name": table_name,
+            "fields": fields_list,
+        }
+        if default_view_name:
+            payload["default_view_name"] = default_view_name
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def delete_tables(
+        self,
+        app_token: str,
+        table_ids: str,
+        table_names: str,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/delete_tables"
+        params = {
+            "app_token": app_token,
+        }
+        if not table_ids:
+            table_id_list = []
+        else:
+            try:
+                table_id_list = json.loads(table_ids)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+
+        if not table_names:
+            table_name_list = []
+        else:
+            try:
+                table_name_list = json.loads(table_names)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+
+        payload = {
+            "table_ids": table_id_list,
+            "table_names": table_name_list,
+        }
+        res = self._send_request(url, params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def list_tables(
+        self,
+        app_token: str,
+        page_token: str,
+        page_size: int = 20,
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/list_tables"
+        params = {
+            "app_token": app_token,
+            "page_token": page_token,
+            "page_size": page_size,
+        }
+        res = self._send_request(url, method="GET", params=params)
+        if "data" in res:
+            return res.get("data")
+        return res
+
+    def read_records(
+        self,
+        app_token: str,
+        table_id: str,
+        table_name: str,
+        record_ids: str,
+        user_id_type: str = "open_id",
+    ) -> dict:
+        url = f"{self.API_BASE_URL}/base/read_records"
+        params = {
+            "app_token": app_token,
+            "table_id": table_id,
+            "table_name": table_name,
+        }
+        if not record_ids:
+            record_id_list = []
+        else:
+            try:
+                record_id_list = json.loads(record_ids)
+            except json.JSONDecodeError:
+                raise ValueError("The input string is not valid JSON")
+        payload = {
+            "record_ids": record_id_list,
+            "user_id_type": user_id_type,
+        }
+        res = self._send_request(url, method="POST", params=params, payload=payload)
+        if "data" in res:
+            return res.get("data")
+        return res

From 0b16270b88a462a28dd2850010e76ed13876ccd2 Mon Sep 17 00:00:00 2001
From: huangyafei <yafei.huangfu@gmail.com>
Date: Thu, 7 Nov 2024 18:11:36 +0800
Subject: [PATCH 04/14] fix typo: Retrieve Chunks API Docs (#10412)

---
 web/app/(commonLayout)/datasets/template/template.en.mdx | 2 +-
 web/app/(commonLayout)/datasets/template/template.zh.mdx | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/web/app/(commonLayout)/datasets/template/template.en.mdx b/web/app/(commonLayout)/datasets/template/template.en.mdx
index 263230d049..02e23429ce 100644
--- a/web/app/(commonLayout)/datasets/template/template.en.mdx
+++ b/web/app/(commonLayout)/datasets/template/template.en.mdx
@@ -1115,7 +1115,7 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
       title="Request"
       tag="POST"
       label="/datasets/{dataset_id}/retrieve"
-      targetCode={`curl --location --request GET '${props.apiBaseUrl}/datasets/{dataset_id}/retrieve' \\\n--header 'Authorization: Bearer {api_key}'\\\n--header 'Content-Type: application/json'\\\n--data-raw '{
+      targetCode={`curl --location --request POST '${props.apiBaseUrl}/datasets/{dataset_id}/retrieve' \\\n--header 'Authorization: Bearer {api_key}'\\\n--header 'Content-Type: application/json'\\\n--data-raw '{
     "query": "test",
     "retrieval_model": {
         "search_method": "keyword_search",
diff --git a/web/app/(commonLayout)/datasets/template/template.zh.mdx b/web/app/(commonLayout)/datasets/template/template.zh.mdx
index 9c25d1e7bb..e5d5f56120 100644
--- a/web/app/(commonLayout)/datasets/template/template.zh.mdx
+++ b/web/app/(commonLayout)/datasets/template/template.zh.mdx
@@ -1116,7 +1116,7 @@ import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from
       title="Request"
       tag="POST"
       label="/datasets/{dataset_id}/retrieve"
-      targetCode={`curl --location --request GET '${props.apiBaseUrl}/datasets/{dataset_id}/retrieve' \\\n--header 'Authorization: Bearer {api_key}'\\\n--header 'Content-Type: application/json'\\\n--data-raw '{
+      targetCode={`curl --location --request POST '${props.apiBaseUrl}/datasets/{dataset_id}/retrieve' \\\n--header 'Authorization: Bearer {api_key}'\\\n--header 'Content-Type: application/json'\\\n--data-raw '{
     "query": "test",
     "retrieval_model": {
         "search_method": "keyword_search",

From dd5ffaf0589406dc051a6dbc7ccc433e2eb9088b Mon Sep 17 00:00:00 2001
From: ice yao <yao3690093@gmail.com>
Date: Thu, 7 Nov 2024 19:31:49 +0800
Subject: [PATCH 05/14] chore: use posixpath to wrapper filepath (#9976)

---
 api/extensions/storage/aliyun_oss_storage.py | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/api/extensions/storage/aliyun_oss_storage.py b/api/extensions/storage/aliyun_oss_storage.py
index 67635b129e..58c917dbd3 100644
--- a/api/extensions/storage/aliyun_oss_storage.py
+++ b/api/extensions/storage/aliyun_oss_storage.py
@@ -1,3 +1,4 @@
+import posixpath
 from collections.abc import Generator
 
 import oss2 as aliyun_s3
@@ -50,9 +51,4 @@ class AliyunOssStorage(BaseStorage):
         self.client.delete_object(self.__wrapper_folder_filename(filename))
 
     def __wrapper_folder_filename(self, filename) -> str:
-        if self.folder:
-            if self.folder.endswith("/"):
-                filename = self.folder + filename
-            else:
-                filename = self.folder + "/" + filename
-        return filename
+        return posixpath.join(self.folder, filename) if self.folder else filename

From 574c4a264f1d3dfca9ecfa72da7a91a3b47df275 Mon Sep 17 00:00:00 2001
From: Bowen Liang <liangbowen@gf.com.cn>
Date: Thu, 7 Nov 2024 21:13:02 +0800
Subject: [PATCH 06/14] chore(lint): Use logging.exception instead of
 logging.error (#10415)

---
 api/controllers/console/datasets/datasets_document.py     | 2 +-
 api/core/app/apps/advanced_chat/generate_task_pipeline.py | 2 +-
 api/core/app/apps/workflow/generate_task_pipeline.py      | 2 +-
 .../model_providers/azure_ai_studio/rerank/rerank.py      | 6 +++---
 api/core/moderation/output_moderation.py                  | 2 +-
 api/core/ops/ops_trace_manager.py                         | 4 ++--
 api/core/rag/datasource/vdb/couchbase/couchbase_vector.py | 2 +-
 api/core/rag/datasource/vdb/lindorm/lindorm_vector.py     | 8 ++++----
 api/core/rag/datasource/vdb/milvus/milvus_vector.py       | 2 +-
 api/core/rag/datasource/vdb/myscale/myscale_vector.py     | 2 +-
 .../rag/datasource/vdb/opensearch/opensearch_vector.py    | 4 ++--
 api/core/rag/embedding/cached_embedding.py                | 2 +-
 api/core/rag/extractor/word_extractor.py                  | 2 +-
 api/core/tools/tool_file_manager.py                       | 2 +-
 api/core/tools/tool_manager.py                            | 2 +-
 api/core/workflow/nodes/base/node.py                      | 2 +-
 .../nodes/question_classifier/question_classifier_node.py | 2 +-
 api/libs/smtp.py                                          | 6 +++---
 api/pyproject.toml                                        | 1 +
 api/services/account_service.py                           | 2 +-
 api/services/tools/api_tools_manage_service.py            | 2 +-
 api/services/tools/tools_transform_service.py             | 2 +-
 22 files changed, 31 insertions(+), 30 deletions(-)

diff --git a/api/controllers/console/datasets/datasets_document.py b/api/controllers/console/datasets/datasets_document.py
index 8e784dc70b..6d6886fcb8 100644
--- a/api/controllers/console/datasets/datasets_document.py
+++ b/api/controllers/console/datasets/datasets_document.py
@@ -945,7 +945,7 @@ class DocumentRetryApi(DocumentResource):
                     raise DocumentAlreadyFinishedError()
                 retry_documents.append(document)
             except Exception as e:
-                logging.error(f"Document {document_id} retry failed: {str(e)}")
+                logging.exception(f"Document {document_id} retry failed: {str(e)}")
                 continue
         # retry document
         DocumentService.retry_document(dataset_id, retry_documents)
diff --git a/api/core/app/apps/advanced_chat/generate_task_pipeline.py b/api/core/app/apps/advanced_chat/generate_task_pipeline.py
index 1fc7ffe2c7..1d4c0ea0fa 100644
--- a/api/core/app/apps/advanced_chat/generate_task_pipeline.py
+++ b/api/core/app/apps/advanced_chat/generate_task_pipeline.py
@@ -242,7 +242,7 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
                     start_listener_time = time.time()
                     yield MessageAudioStreamResponse(audio=audio_trunk.audio, task_id=task_id)
             except Exception as e:
-                logger.error(e)
+                logger.exception(e)
                 break
         if tts_publisher:
             yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
diff --git a/api/core/app/apps/workflow/generate_task_pipeline.py b/api/core/app/apps/workflow/generate_task_pipeline.py
index d119d94a61..aaa4824fe8 100644
--- a/api/core/app/apps/workflow/generate_task_pipeline.py
+++ b/api/core/app/apps/workflow/generate_task_pipeline.py
@@ -216,7 +216,7 @@ class WorkflowAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCycleMa
                 else:
                     yield MessageAudioStreamResponse(audio=audio_trunk.audio, task_id=task_id)
             except Exception as e:
-                logger.error(e)
+                logger.exception(e)
                 break
         if tts_publisher:
             yield MessageAudioEndStreamResponse(audio="", task_id=task_id)
diff --git a/api/core/model_runtime/model_providers/azure_ai_studio/rerank/rerank.py b/api/core/model_runtime/model_providers/azure_ai_studio/rerank/rerank.py
index b6dfb20b66..84672520e0 100644
--- a/api/core/model_runtime/model_providers/azure_ai_studio/rerank/rerank.py
+++ b/api/core/model_runtime/model_providers/azure_ai_studio/rerank/rerank.py
@@ -47,9 +47,9 @@ class AzureRerankModel(RerankModel):
                 result = response.read()
                 return json.loads(result)
         except urllib.error.HTTPError as error:
-            logger.error(f"The request failed with status code: {error.code}")
-            logger.error(error.info())
-            logger.error(error.read().decode("utf8", "ignore"))
+            logger.exception(f"The request failed with status code: {error.code}")
+            logger.exception(error.info())
+            logger.exception(error.read().decode("utf8", "ignore"))
             raise
 
     def _invoke(
diff --git a/api/core/moderation/output_moderation.py b/api/core/moderation/output_moderation.py
index d8d794be18..83f4d2d57d 100644
--- a/api/core/moderation/output_moderation.py
+++ b/api/core/moderation/output_moderation.py
@@ -126,6 +126,6 @@ class OutputModeration(BaseModel):
             result: ModerationOutputsResult = moderation_factory.moderation_for_outputs(moderation_buffer)
             return result
         except Exception as e:
-            logger.error("Moderation Output error: %s", e)
+            logger.exception("Moderation Output error: %s", e)
 
         return None
diff --git a/api/core/ops/ops_trace_manager.py b/api/core/ops/ops_trace_manager.py
index 764944f799..986749f056 100644
--- a/api/core/ops/ops_trace_manager.py
+++ b/api/core/ops/ops_trace_manager.py
@@ -708,7 +708,7 @@ class TraceQueueManager:
                 trace_task.app_id = self.app_id
                 trace_manager_queue.put(trace_task)
         except Exception as e:
-            logging.error(f"Error adding trace task: {e}")
+            logging.exception(f"Error adding trace task: {e}")
         finally:
             self.start_timer()
 
@@ -727,7 +727,7 @@ class TraceQueueManager:
             if tasks:
                 self.send_to_celery(tasks)
         except Exception as e:
-            logging.error(f"Error processing trace tasks: {e}")
+            logging.exception(f"Error processing trace tasks: {e}")
 
     def start_timer(self):
         global trace_manager_timer
diff --git a/api/core/rag/datasource/vdb/couchbase/couchbase_vector.py b/api/core/rag/datasource/vdb/couchbase/couchbase_vector.py
index 3f88d2ca2b..98da5e3d5e 100644
--- a/api/core/rag/datasource/vdb/couchbase/couchbase_vector.py
+++ b/api/core/rag/datasource/vdb/couchbase/couchbase_vector.py
@@ -242,7 +242,7 @@ class CouchbaseVector(BaseVector):
         try:
             self._cluster.query(query, named_parameters={"doc_ids": ids}).execute()
         except Exception as e:
-            logger.error(e)
+            logger.exception(e)
 
     def delete_by_document_id(self, document_id: str):
         query = f"""
diff --git a/api/core/rag/datasource/vdb/lindorm/lindorm_vector.py b/api/core/rag/datasource/vdb/lindorm/lindorm_vector.py
index abd8261a69..30d7f09ec2 100644
--- a/api/core/rag/datasource/vdb/lindorm/lindorm_vector.py
+++ b/api/core/rag/datasource/vdb/lindorm/lindorm_vector.py
@@ -79,7 +79,7 @@ class LindormVectorStore(BaseVector):
                 existing_docs = self._client.mget(index=self._collection_name, body={"ids": batch_ids}, _source=False)
                 return {doc["_id"] for doc in existing_docs["docs"] if doc["found"]}
             except Exception as e:
-                logger.error(f"Error fetching batch {batch_ids}: {e}")
+                logger.exception(f"Error fetching batch {batch_ids}: {e}")
                 return set()
 
         @retry(stop=stop_after_attempt(3), wait=wait_fixed(60))
@@ -96,7 +96,7 @@ class LindormVectorStore(BaseVector):
                 )
                 return {doc["_id"] for doc in existing_docs["docs"] if doc["found"]}
             except Exception as e:
-                logger.error(f"Error fetching batch {batch_ids}: {e}")
+                logger.exception(f"Error fetching batch {batch_ids}: {e}")
                 return set()
 
         if ids is None:
@@ -177,7 +177,7 @@ class LindormVectorStore(BaseVector):
             else:
                 logger.warning(f"Index '{self._collection_name}' does not exist. No deletion performed.")
         except Exception as e:
-            logger.error(f"Error occurred while deleting the index: {e}")
+            logger.exception(f"Error occurred while deleting the index: {e}")
             raise e
 
     def text_exists(self, id: str) -> bool:
@@ -201,7 +201,7 @@ class LindormVectorStore(BaseVector):
         try:
             response = self._client.search(index=self._collection_name, body=query)
         except Exception as e:
-            logger.error(f"Error executing search: {e}")
+            logger.exception(f"Error executing search: {e}")
             raise
 
         docs_and_scores = []
diff --git a/api/core/rag/datasource/vdb/milvus/milvus_vector.py b/api/core/rag/datasource/vdb/milvus/milvus_vector.py
index 080a1ef567..5a263d6e78 100644
--- a/api/core/rag/datasource/vdb/milvus/milvus_vector.py
+++ b/api/core/rag/datasource/vdb/milvus/milvus_vector.py
@@ -86,7 +86,7 @@ class MilvusVector(BaseVector):
                 ids = self._client.insert(collection_name=self._collection_name, data=batch_insert_list)
                 pks.extend(ids)
             except MilvusException as e:
-                logger.error("Failed to insert batch starting at entity: %s/%s", i, total_count)
+                logger.exception("Failed to insert batch starting at entity: %s/%s", i, total_count)
                 raise e
         return pks
 
diff --git a/api/core/rag/datasource/vdb/myscale/myscale_vector.py b/api/core/rag/datasource/vdb/myscale/myscale_vector.py
index 1fca926a2d..2610b60a77 100644
--- a/api/core/rag/datasource/vdb/myscale/myscale_vector.py
+++ b/api/core/rag/datasource/vdb/myscale/myscale_vector.py
@@ -142,7 +142,7 @@ class MyScaleVector(BaseVector):
                 for r in self._client.query(sql).named_results()
             ]
         except Exception as e:
-            logging.error(f"\033[91m\033[1m{type(e)}\033[0m \033[95m{str(e)}\033[0m")
+            logging.exception(f"\033[91m\033[1m{type(e)}\033[0m \033[95m{str(e)}\033[0m")
             return []
 
     def delete(self) -> None:
diff --git a/api/core/rag/datasource/vdb/opensearch/opensearch_vector.py b/api/core/rag/datasource/vdb/opensearch/opensearch_vector.py
index 0e0f107268..49eb00f140 100644
--- a/api/core/rag/datasource/vdb/opensearch/opensearch_vector.py
+++ b/api/core/rag/datasource/vdb/opensearch/opensearch_vector.py
@@ -129,7 +129,7 @@ class OpenSearchVector(BaseVector):
                     if status == 404:
                         logger.warning(f"Document not found for deletion: {doc_id}")
                     else:
-                        logger.error(f"Error deleting document: {error}")
+                        logger.exception(f"Error deleting document: {error}")
 
     def delete(self) -> None:
         self._client.indices.delete(index=self._collection_name.lower())
@@ -158,7 +158,7 @@ class OpenSearchVector(BaseVector):
         try:
             response = self._client.search(index=self._collection_name.lower(), body=query)
         except Exception as e:
-            logger.error(f"Error executing search: {e}")
+            logger.exception(f"Error executing search: {e}")
             raise
 
         docs = []
diff --git a/api/core/rag/embedding/cached_embedding.py b/api/core/rag/embedding/cached_embedding.py
index b3e93ce760..3ac65b88bb 100644
--- a/api/core/rag/embedding/cached_embedding.py
+++ b/api/core/rag/embedding/cached_embedding.py
@@ -89,7 +89,7 @@ class CacheEmbedding(Embeddings):
                     db.session.rollback()
             except Exception as ex:
                 db.session.rollback()
-                logger.error("Failed to embed documents: %s", ex)
+                logger.exception("Failed to embed documents: %s", ex)
                 raise ex
 
         return text_embeddings
diff --git a/api/core/rag/extractor/word_extractor.py b/api/core/rag/extractor/word_extractor.py
index d4434ea28f..576bba0142 100644
--- a/api/core/rag/extractor/word_extractor.py
+++ b/api/core/rag/extractor/word_extractor.py
@@ -230,7 +230,7 @@ class WordExtractor(BaseExtractor):
                                 for i in url_pattern.findall(x.text):
                                     hyperlinks_url = str(i)
                     except Exception as e:
-                        logger.error(e)
+                        logger.exception(e)
 
         def parse_paragraph(paragraph):
             paragraph_content = []
diff --git a/api/core/tools/tool_file_manager.py b/api/core/tools/tool_file_manager.py
index 1a28df31bc..ff56e20e87 100644
--- a/api/core/tools/tool_file_manager.py
+++ b/api/core/tools/tool_file_manager.py
@@ -98,7 +98,7 @@ class ToolFileManager:
             response.raise_for_status()
             blob = response.content
         except Exception as e:
-            logger.error(f"Failed to download file from {file_url}: {e}")
+            logger.exception(f"Failed to download file from {file_url}: {e}")
             raise
 
         mimetype = guess_type(file_url)[0] or "octet/stream"
diff --git a/api/core/tools/tool_manager.py b/api/core/tools/tool_manager.py
index 6abe0a9cba..bf2ad13620 100644
--- a/api/core/tools/tool_manager.py
+++ b/api/core/tools/tool_manager.py
@@ -388,7 +388,7 @@ class ToolManager:
                     yield provider
 
                 except Exception as e:
-                    logger.error(f"load builtin provider {provider} error: {e}")
+                    logger.exception(f"load builtin provider {provider} error: {e}")
                     continue
         # set builtin providers loaded
         cls._builtin_providers_loaded = True
diff --git a/api/core/workflow/nodes/base/node.py b/api/core/workflow/nodes/base/node.py
index 053a339ba7..1433c8eaed 100644
--- a/api/core/workflow/nodes/base/node.py
+++ b/api/core/workflow/nodes/base/node.py
@@ -69,7 +69,7 @@ class BaseNode(Generic[GenericNodeData]):
         try:
             result = self._run()
         except Exception as e:
-            logger.error(f"Node {self.node_id} failed to run: {e}")
+            logger.exception(f"Node {self.node_id} failed to run: {e}")
             result = NodeRunResult(
                 status=WorkflowNodeExecutionStatus.FAILED,
                 error=str(e),
diff --git a/api/core/workflow/nodes/question_classifier/question_classifier_node.py b/api/core/workflow/nodes/question_classifier/question_classifier_node.py
index 0489020e5e..744dfd3d8d 100644
--- a/api/core/workflow/nodes/question_classifier/question_classifier_node.py
+++ b/api/core/workflow/nodes/question_classifier/question_classifier_node.py
@@ -127,7 +127,7 @@ class QuestionClassifierNode(LLMNode):
                     category_id = category_id_result
 
         except OutputParserError:
-            logging.error(f"Failed to parse result text: {result_text}")
+            logging.exception(f"Failed to parse result text: {result_text}")
         try:
             process_data = {
                 "model_mode": model_config.mode,
diff --git a/api/libs/smtp.py b/api/libs/smtp.py
index bd7de7dd68..d57d99f3b7 100644
--- a/api/libs/smtp.py
+++ b/api/libs/smtp.py
@@ -39,13 +39,13 @@ class SMTPClient:
 
             smtp.sendmail(self._from, mail["to"], msg.as_string())
         except smtplib.SMTPException as e:
-            logging.error(f"SMTP error occurred: {str(e)}")
+            logging.exception(f"SMTP error occurred: {str(e)}")
             raise
         except TimeoutError as e:
-            logging.error(f"Timeout occurred while sending email: {str(e)}")
+            logging.exception(f"Timeout occurred while sending email: {str(e)}")
             raise
         except Exception as e:
-            logging.error(f"Unexpected error occurred while sending email: {str(e)}")
+            logging.exception(f"Unexpected error occurred while sending email: {str(e)}")
             raise
         finally:
             if smtp:
diff --git a/api/pyproject.toml b/api/pyproject.toml
index 4438cf61db..928dee975b 100644
--- a/api/pyproject.toml
+++ b/api/pyproject.toml
@@ -34,6 +34,7 @@ select = [
     "RUF101", # redirected-noqa
     "S506", # unsafe-yaml-load
     "SIM", # flake8-simplify rules
+    "TRY400", # error-instead-of-exception
     "UP", # pyupgrade rules
     "W191", # tab-indentation
     "W605", # invalid-escape-sequence
diff --git a/api/services/account_service.py b/api/services/account_service.py
index dceca06185..963a055948 100644
--- a/api/services/account_service.py
+++ b/api/services/account_service.py
@@ -821,7 +821,7 @@ class RegisterService:
             db.session.rollback()
         except Exception as e:
             db.session.rollback()
-            logging.error(f"Register failed: {e}")
+            logging.exception(f"Register failed: {e}")
             raise AccountRegisterError(f"Registration failed: {e}") from e
 
         return account
diff --git a/api/services/tools/api_tools_manage_service.py b/api/services/tools/api_tools_manage_service.py
index 257c6cf52b..4a93891855 100644
--- a/api/services/tools/api_tools_manage_service.py
+++ b/api/services/tools/api_tools_manage_service.py
@@ -193,7 +193,7 @@ class ApiToolManageService:
             # try to parse schema, avoid SSRF attack
             ApiToolManageService.parser_api_schema(schema)
         except Exception as e:
-            logger.error(f"parse api schema error: {str(e)}")
+            logger.exception(f"parse api schema error: {str(e)}")
             raise ValueError("invalid schema, please check the url you provided")
 
         return {"schema": schema}
diff --git a/api/services/tools/tools_transform_service.py b/api/services/tools/tools_transform_service.py
index 4af73d5063..e535ddb575 100644
--- a/api/services/tools/tools_transform_service.py
+++ b/api/services/tools/tools_transform_service.py
@@ -183,7 +183,7 @@ class ToolTransformService:
         try:
             username = db_provider.user.name
         except Exception as e:
-            logger.error(f"failed to get user name for api provider {db_provider.id}: {str(e)}")
+            logger.exception(f"failed to get user name for api provider {db_provider.id}: {str(e)}")
         # add provider into providers
         credentials = db_provider.credentials
         result = UserToolProvider(

From 6e23903c635f5cd1f1009f6994db9cdef6b060aa Mon Sep 17 00:00:00 2001
From: Benjamin <benjaminx@gmail.com>
Date: Thu, 7 Nov 2024 21:13:23 +0800
Subject: [PATCH 07/14] Conversation delete issue (#10423)

---
 api/controllers/service_api/app/conversation.py | 8 ++++++--
 api/fields/conversation_fields.py               | 4 ++++
 api/services/conversation_service.py            | 2 --
 3 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/api/controllers/service_api/app/conversation.py b/api/controllers/service_api/app/conversation.py
index a0bd1b7412..815fd6a27a 100644
--- a/api/controllers/service_api/app/conversation.py
+++ b/api/controllers/service_api/app/conversation.py
@@ -7,7 +7,11 @@ from controllers.service_api import api
 from controllers.service_api.app.error import NotChatAppError
 from controllers.service_api.wraps import FetchUserArg, WhereisUserArg, validate_app_token
 from core.app.entities.app_invoke_entities import InvokeFrom
-from fields.conversation_fields import conversation_infinite_scroll_pagination_fields, simple_conversation_fields
+from fields.conversation_fields import (
+    conversation_delete_fields,
+    conversation_infinite_scroll_pagination_fields,
+    simple_conversation_fields,
+)
 from libs.helper import uuid_value
 from models.model import App, AppMode, EndUser
 from services.conversation_service import ConversationService
@@ -49,7 +53,7 @@ class ConversationApi(Resource):
 
 class ConversationDetailApi(Resource):
     @validate_app_token(fetch_user_arg=FetchUserArg(fetch_from=WhereisUserArg.JSON))
-    @marshal_with(simple_conversation_fields)
+    @marshal_with(conversation_delete_fields)
     def delete(self, app_model: App, end_user: EndUser, c_id):
         app_mode = AppMode.value_of(app_model.mode)
         if app_mode not in {AppMode.CHAT, AppMode.AGENT_CHAT, AppMode.ADVANCED_CHAT}:
diff --git a/api/fields/conversation_fields.py b/api/fields/conversation_fields.py
index 2eb19c2667..5bd21be807 100644
--- a/api/fields/conversation_fields.py
+++ b/api/fields/conversation_fields.py
@@ -202,6 +202,10 @@ simple_conversation_fields = {
     "updated_at": TimestampField,
 }
 
+conversation_delete_fields = {
+    "result": fields.String,
+}
+
 conversation_infinite_scroll_pagination_fields = {
     "limit": fields.Integer,
     "has_more": fields.Boolean,
diff --git a/api/services/conversation_service.py b/api/services/conversation_service.py
index ac9f577c32..f9e41988c0 100644
--- a/api/services/conversation_service.py
+++ b/api/services/conversation_service.py
@@ -162,5 +162,3 @@ class ConversationService:
         conversation.is_deleted = True
         conversation.updated_at = datetime.now(timezone.utc).replace(tzinfo=None)
         db.session.commit()
-
-        return {"result": "success"}, 200

From 98b3e37144a4437cb20f339fe092bab742b4de3b Mon Sep 17 00:00:00 2001
From: katsuma <71871253+SuzukiKatsuma@users.noreply.github.com>
Date: Thu, 7 Nov 2024 22:21:50 +0900
Subject: [PATCH 08/14] fix: simplify Enter key handling and remove unused ref
 (#10413)

---
 .../base/chat/chat/chat-input-area/index.tsx   | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)

diff --git a/web/app/components/base/chat/chat/chat-input-area/index.tsx b/web/app/components/base/chat/chat/chat-input-area/index.tsx
index 05aaaa6bc2..32d841148a 100644
--- a/web/app/components/base/chat/chat/chat-input-area/index.tsx
+++ b/web/app/components/base/chat/chat/chat-input-area/index.tsx
@@ -1,6 +1,5 @@
 import {
   useCallback,
-  useRef,
   useState,
 } from 'react'
 import Textarea from 'rc-textarea'
@@ -63,7 +62,6 @@ const ChatInputArea = ({
     isMultipleLine,
   } = useTextAreaHeight()
   const [query, setQuery] = useState('')
-  const isUseInputMethod = useRef(false)
   const [showVoiceInput, setShowVoiceInput] = useState(false)
   const filesStore = useFileStore()
   const {
@@ -95,20 +93,11 @@ const ChatInputArea = ({
     }
   }
 
-  const handleKeyUp = (e: React.KeyboardEvent<HTMLTextAreaElement>) => {
-    if (e.key === 'Enter') {
-      e.preventDefault()
-      // prevent send message when using input method enter
-      if (!e.shiftKey && !isUseInputMethod.current)
-        handleSend()
-    }
-  }
-
   const handleKeyDown = (e: React.KeyboardEvent<HTMLTextAreaElement>) => {
-    isUseInputMethod.current = e.nativeEvent.isComposing
-    if (e.key === 'Enter' && !e.shiftKey) {
-      setQuery(query.replace(/\n$/, ''))
+    if (e.key === 'Enter' && !e.shiftKey && !e.nativeEvent.isComposing) {
       e.preventDefault()
+      setQuery(query.replace(/\n$/, ''))
+      handleSend()
     }
   }
 
@@ -165,7 +154,6 @@ const ChatInputArea = ({
                   setQuery(e.target.value)
                   handleTextareaResize()
                 }}
-                onKeyUp={handleKeyUp}
                 onKeyDown={handleKeyDown}
                 onPaste={handleClipboardPasteFile}
                 onDragEnter={handleDragFileEnter}

From a60133bfb329ca1fd67634ab4e6099885008fd6c Mon Sep 17 00:00:00 2001
From: Bowen Liang <liangbowen@gf.com.cn>
Date: Fri, 8 Nov 2024 09:33:12 +0800
Subject: [PATCH 09/14] fix: config violations when running db migtration ci
 tests (#10428)

---
 .github/workflows/db-migration-test.yml |  1 +
 api/.env.example                        |  8 +++----
 api/configs/feature/__init__.py         | 10 +++++++++
 api/core/helper/ssrf_proxy.py           | 30 ++++++++++---------------
 4 files changed, 27 insertions(+), 22 deletions(-)

diff --git a/.github/workflows/db-migration-test.yml b/.github/workflows/db-migration-test.yml
index b8246aacb3..c6fe87264d 100644
--- a/.github/workflows/db-migration-test.yml
+++ b/.github/workflows/db-migration-test.yml
@@ -6,6 +6,7 @@ on:
       - main
     paths:
       - api/migrations/**
+      - .github/workflows/db-migration-test.yml
 
 concurrency:
   group: db-migration-test-${{ github.ref }}
diff --git a/api/.env.example b/api/.env.example
index a92490608f..1214154389 100644
--- a/api/.env.example
+++ b/api/.env.example
@@ -324,10 +324,10 @@ UNSTRUCTURED_API_KEY=
 SSRF_PROXY_HTTP_URL=
 SSRF_PROXY_HTTPS_URL=
 SSRF_DEFAULT_MAX_RETRIES=3
-SSRF_DEFAULT_TIME_OUT=
-SSRF_DEFAULT_CONNECT_TIME_OUT=
-SSRF_DEFAULT_READ_TIME_OUT=
-SSRF_DEFAULT_WRITE_TIME_OUT=
+SSRF_DEFAULT_TIME_OUT=5
+SSRF_DEFAULT_CONNECT_TIME_OUT=5
+SSRF_DEFAULT_READ_TIME_OUT=5
+SSRF_DEFAULT_WRITE_TIME_OUT=5
 
 BATCH_UPLOAD_LIMIT=10
 KEYWORD_DATA_SOURCE_TYPE=database
diff --git a/api/configs/feature/__init__.py b/api/configs/feature/__init__.py
index 3ac2c28c1f..5babae8810 100644
--- a/api/configs/feature/__init__.py
+++ b/api/configs/feature/__init__.py
@@ -276,6 +276,16 @@ class HttpConfig(BaseSettings):
         default=1 * 1024 * 1024,
     )
 
+    SSRF_DEFAULT_MAX_RETRIES: PositiveInt = Field(
+        description="Maximum number of retries for network requests (SSRF)",
+        default=3,
+    )
+
+    SSRF_PROXY_ALL_URL: Optional[str] = Field(
+        description="Proxy URL for HTTP or HTTPS requests to prevent Server-Side Request Forgery (SSRF)",
+        default=None,
+    )
+
     SSRF_PROXY_HTTP_URL: Optional[str] = Field(
         description="Proxy URL for HTTP requests to prevent Server-Side Request Forgery (SSRF)",
         default=None,
diff --git a/api/core/helper/ssrf_proxy.py b/api/core/helper/ssrf_proxy.py
index df812ca83f..374bd9d57b 100644
--- a/api/core/helper/ssrf_proxy.py
+++ b/api/core/helper/ssrf_proxy.py
@@ -3,26 +3,20 @@ Proxy requests to avoid SSRF
 """
 
 import logging
-import os
 import time
 
 import httpx
 
-SSRF_PROXY_ALL_URL = os.getenv("SSRF_PROXY_ALL_URL", "")
-SSRF_PROXY_HTTP_URL = os.getenv("SSRF_PROXY_HTTP_URL", "")
-SSRF_PROXY_HTTPS_URL = os.getenv("SSRF_PROXY_HTTPS_URL", "")
-SSRF_DEFAULT_MAX_RETRIES = int(os.getenv("SSRF_DEFAULT_MAX_RETRIES", "3"))
-SSRF_DEFAULT_TIME_OUT = float(os.getenv("SSRF_DEFAULT_TIME_OUT", "5"))
-SSRF_DEFAULT_CONNECT_TIME_OUT = float(os.getenv("SSRF_DEFAULT_CONNECT_TIME_OUT", "5"))
-SSRF_DEFAULT_READ_TIME_OUT = float(os.getenv("SSRF_DEFAULT_READ_TIME_OUT", "5"))
-SSRF_DEFAULT_WRITE_TIME_OUT = float(os.getenv("SSRF_DEFAULT_WRITE_TIME_OUT", "5"))
+from configs import dify_config
+
+SSRF_DEFAULT_MAX_RETRIES = dify_config.SSRF_DEFAULT_MAX_RETRIES
 
 proxy_mounts = (
     {
-        "http://": httpx.HTTPTransport(proxy=SSRF_PROXY_HTTP_URL),
-        "https://": httpx.HTTPTransport(proxy=SSRF_PROXY_HTTPS_URL),
+        "http://": httpx.HTTPTransport(proxy=dify_config.SSRF_PROXY_HTTP_URL),
+        "https://": httpx.HTTPTransport(proxy=dify_config.SSRF_PROXY_HTTPS_URL),
     }
-    if SSRF_PROXY_HTTP_URL and SSRF_PROXY_HTTPS_URL
+    if dify_config.SSRF_PROXY_HTTP_URL and dify_config.SSRF_PROXY_HTTPS_URL
     else None
 )
 
@@ -38,17 +32,17 @@ def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
 
     if "timeout" not in kwargs:
         kwargs["timeout"] = httpx.Timeout(
-            SSRF_DEFAULT_TIME_OUT,
-            connect=SSRF_DEFAULT_CONNECT_TIME_OUT,
-            read=SSRF_DEFAULT_READ_TIME_OUT,
-            write=SSRF_DEFAULT_WRITE_TIME_OUT,
+            timeout=dify_config.SSRF_DEFAULT_TIME_OUT,
+            connect=dify_config.SSRF_DEFAULT_CONNECT_TIME_OUT,
+            read=dify_config.SSRF_DEFAULT_READ_TIME_OUT,
+            write=dify_config.SSRF_DEFAULT_WRITE_TIME_OUT,
         )
 
     retries = 0
     while retries <= max_retries:
         try:
-            if SSRF_PROXY_ALL_URL:
-                with httpx.Client(proxy=SSRF_PROXY_ALL_URL) as client:
+            if dify_config.SSRF_PROXY_ALL_URL:
+                with httpx.Client(proxy=dify_config.SSRF_PROXY_ALL_URL) as client:
                     response = client.request(method=method, url=url, **kwargs)
             elif proxy_mounts:
                 with httpx.Client(mounts=proxy_mounts) as client:

From 438ad8148b7e1cdf81ec4803cd2fa08e4da15d21 Mon Sep 17 00:00:00 2001
From: -LAN- <laipz8200@outlook.com>
Date: Fri, 8 Nov 2024 09:33:40 +0800
Subject: [PATCH 10/14] fix(http_request): send form data (#10431)

---
 .../workflow/nodes/http_request/executor.py   | 22 +++---
 .../test_http_request_executor.py             | 69 +++++++++++++++++++
 2 files changed, 78 insertions(+), 13 deletions(-)

diff --git a/api/core/workflow/nodes/http_request/executor.py b/api/core/workflow/nodes/http_request/executor.py
index d90dfcc766..80b322b068 100644
--- a/api/core/workflow/nodes/http_request/executor.py
+++ b/api/core/workflow/nodes/http_request/executor.py
@@ -97,15 +97,6 @@ class Executor:
         headers = self.variable_pool.convert_template(self.node_data.headers).text
         self.headers = _plain_text_to_dict(headers)
 
-        body = self.node_data.body
-        if body is None:
-            return
-        if "content-type" not in (k.lower() for k in self.headers) and body.type in BODY_TYPE_TO_CONTENT_TYPE:
-            self.headers["Content-Type"] = BODY_TYPE_TO_CONTENT_TYPE[body.type]
-        if body.type == "form-data":
-            self.boundary = f"----WebKitFormBoundary{_generate_random_string(16)}"
-            self.headers["Content-Type"] = f"multipart/form-data; boundary={self.boundary}"
-
     def _init_body(self):
         body = self.node_data.body
         if body is not None:
@@ -154,9 +145,8 @@ class Executor:
                         for k, v in files.items()
                         if v.related_id is not None
                     }
-
                     self.data = form_data
-                    self.files = files
+                    self.files = files or None
 
     def _assembling_headers(self) -> dict[str, Any]:
         authorization = deepcopy(self.auth)
@@ -217,6 +207,7 @@ class Executor:
             "timeout": (self.timeout.connect, self.timeout.read, self.timeout.write),
             "follow_redirects": True,
         }
+        # request_args = {k: v for k, v in request_args.items() if v is not None}
 
         response = getattr(ssrf_proxy, self.method)(**request_args)
         return response
@@ -244,6 +235,13 @@ class Executor:
         raw += f"Host: {url_parts.netloc}\r\n"
 
         headers = self._assembling_headers()
+        body = self.node_data.body
+        boundary = f"----WebKitFormBoundary{_generate_random_string(16)}"
+        if body:
+            if "content-type" not in (k.lower() for k in self.headers) and body.type in BODY_TYPE_TO_CONTENT_TYPE:
+                headers["Content-Type"] = BODY_TYPE_TO_CONTENT_TYPE[body.type]
+            if body.type == "form-data":
+                headers["Content-Type"] = f"multipart/form-data; boundary={boundary}"
         for k, v in headers.items():
             if self.auth.type == "api-key":
                 authorization_header = "Authorization"
@@ -256,7 +254,6 @@ class Executor:
 
         body = ""
         if self.files:
-            boundary = self.boundary
             for k, v in self.files.items():
                 body += f"--{boundary}\r\n"
                 body += f'Content-Disposition: form-data; name="{k}"\r\n\r\n'
@@ -271,7 +268,6 @@ class Executor:
             elif self.data and self.node_data.body.type == "x-www-form-urlencoded":
                 body = urlencode(self.data)
             elif self.data and self.node_data.body.type == "form-data":
-                boundary = self.boundary
                 for key, value in self.data.items():
                     body += f"--{boundary}\r\n"
                     body += f'Content-Disposition: form-data; name="{key}"\r\n\r\n'
diff --git a/api/tests/unit_tests/core/workflow/nodes/http_request/test_http_request_executor.py b/api/tests/unit_tests/core/workflow/nodes/http_request/test_http_request_executor.py
index 12c469a81a..7c19de6078 100644
--- a/api/tests/unit_tests/core/workflow/nodes/http_request/test_http_request_executor.py
+++ b/api/tests/unit_tests/core/workflow/nodes/http_request/test_http_request_executor.py
@@ -196,3 +196,72 @@ def test_extract_selectors_from_template_with_newline():
     )
 
     assert executor.params == {"test": "line1\nline2"}
+
+
+def test_executor_with_form_data():
+    # Prepare the variable pool
+    variable_pool = VariablePool(
+        system_variables={},
+        user_inputs={},
+    )
+    variable_pool.add(["pre_node_id", "text_field"], "Hello, World!")
+    variable_pool.add(["pre_node_id", "number_field"], 42)
+
+    # Prepare the node data
+    node_data = HttpRequestNodeData(
+        title="Test Form Data",
+        method="post",
+        url="https://api.example.com/upload",
+        authorization=HttpRequestNodeAuthorization(type="no-auth"),
+        headers="Content-Type: multipart/form-data",
+        params="",
+        body=HttpRequestNodeBody(
+            type="form-data",
+            data=[
+                BodyData(
+                    key="text_field",
+                    type="text",
+                    value="{{#pre_node_id.text_field#}}",
+                ),
+                BodyData(
+                    key="number_field",
+                    type="text",
+                    value="{{#pre_node_id.number_field#}}",
+                ),
+            ],
+        ),
+    )
+
+    # Initialize the Executor
+    executor = Executor(
+        node_data=node_data,
+        timeout=HttpRequestNodeTimeout(connect=10, read=30, write=30),
+        variable_pool=variable_pool,
+    )
+
+    # Check the executor's data
+    assert executor.method == "post"
+    assert executor.url == "https://api.example.com/upload"
+    assert "Content-Type" in executor.headers
+    assert "multipart/form-data" in executor.headers["Content-Type"]
+    assert executor.params == {}
+    assert executor.json is None
+    assert executor.files is None
+    assert executor.content is None
+
+    # Check that the form data is correctly loaded in executor.data
+    assert isinstance(executor.data, dict)
+    assert "text_field" in executor.data
+    assert executor.data["text_field"] == "Hello, World!"
+    assert "number_field" in executor.data
+    assert executor.data["number_field"] == "42"
+
+    # Check the raw request (to_log method)
+    raw_request = executor.to_log()
+    assert "POST /upload HTTP/1.1" in raw_request
+    assert "Host: api.example.com" in raw_request
+    assert "Content-Type: multipart/form-data" in raw_request
+    assert "text_field" in raw_request
+    assert "Hello, World!" in raw_request
+    assert "number_field" in raw_request
+    assert "42" in raw_request

From 0ebe198ff102825b8ed07aeccc656c4b4e9c1260 Mon Sep 17 00:00:00 2001
From: Bowen Liang <liangbowen@gf.com.cn>
Date: Fri, 8 Nov 2024 09:34:11 +0800
Subject: [PATCH 11/14] chore: use DEBUG in dify_config instead of parsing raw
 system environment variable in place (#10437)

---
 api/app.py                                       | 2 +-
 api/app_factory.py                               | 4 +++-
 api/core/app/apps/advanced_chat/app_generator.py | 4 ++--
 api/core/app/apps/agent_chat/app_generator.py    | 4 ++--
 api/core/app/apps/chat/app_generator.py          | 4 ++--
 api/core/app/apps/completion/app_generator.py    | 4 ++--
 api/core/app/apps/workflow/app_generator.py      | 4 ++--
 api/core/model_manager.py                        | 4 ++--
 8 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/api/app.py b/api/app.py
index ed214bde97..93774bd75d 100644
--- a/api/app.py
+++ b/api/app.py
@@ -2,7 +2,7 @@ import os
 
 from configs import dify_config
 
-if os.environ.get("DEBUG", "false").lower() != "true":
+if dify_config.DEBUG:
     from gevent import monkey
 
     monkey.patch_all()
diff --git a/api/app_factory.py b/api/app_factory.py
index aba78ccab8..f53cf14f4b 100644
--- a/api/app_factory.py
+++ b/api/app_factory.py
@@ -1,6 +1,8 @@
 import os
 
-if os.environ.get("DEBUG", "false").lower() != "true":
+from configs import dify_config
+
+if dify_config.DEBUG:
     from gevent import monkey
 
     monkey.patch_all()
diff --git a/api/core/app/apps/advanced_chat/app_generator.py b/api/core/app/apps/advanced_chat/app_generator.py
index 39ab87c914..3010f8a03f 100644
--- a/api/core/app/apps/advanced_chat/app_generator.py
+++ b/api/core/app/apps/advanced_chat/app_generator.py
@@ -1,6 +1,5 @@
 import contextvars
 import logging
-import os
 import threading
 import uuid
 from collections.abc import Generator
@@ -10,6 +9,7 @@ from flask import Flask, current_app
 from pydantic import ValidationError
 
 import contexts
+from configs import dify_config
 from constants import UUID_NIL
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.advanced_chat.app_config_manager import AdvancedChatAppConfigManager
@@ -317,7 +317,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
                 logger.exception("Validation Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except (ValueError, InvokeError) as e:
-                if os.environ.get("DEBUG", "false").lower() == "true":
+                if dify_config.DEBUG:
                     logger.exception("Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except Exception as e:
diff --git a/api/core/app/apps/agent_chat/app_generator.py b/api/core/app/apps/agent_chat/app_generator.py
index de12f5a441..73d433d94d 100644
--- a/api/core/app/apps/agent_chat/app_generator.py
+++ b/api/core/app/apps/agent_chat/app_generator.py
@@ -1,5 +1,4 @@
 import logging
-import os
 import threading
 import uuid
 from collections.abc import Generator
@@ -8,6 +7,7 @@ from typing import Any, Literal, Union, overload
 from flask import Flask, current_app
 from pydantic import ValidationError
 
+from configs import dify_config
 from constants import UUID_NIL
 from core.app.app_config.easy_ui_based_app.model_config.converter import ModelConfigConverter
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
@@ -230,7 +230,7 @@ class AgentChatAppGenerator(MessageBasedAppGenerator):
                 logger.exception("Validation Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except (ValueError, InvokeError) as e:
-                if os.environ.get("DEBUG") and os.environ.get("DEBUG").lower() == "true":
+                if dify_config.DEBUG:
                     logger.exception("Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except Exception as e:
diff --git a/api/core/app/apps/chat/app_generator.py b/api/core/app/apps/chat/app_generator.py
index 5c074f5306..d0ba90cc5e 100644
--- a/api/core/app/apps/chat/app_generator.py
+++ b/api/core/app/apps/chat/app_generator.py
@@ -1,5 +1,4 @@
 import logging
-import os
 import threading
 import uuid
 from collections.abc import Generator
@@ -8,6 +7,7 @@ from typing import Any, Literal, Union, overload
 from flask import Flask, current_app
 from pydantic import ValidationError
 
+from configs import dify_config
 from constants import UUID_NIL
 from core.app.app_config.easy_ui_based_app.model_config.converter import ModelConfigConverter
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
@@ -227,7 +227,7 @@ class ChatAppGenerator(MessageBasedAppGenerator):
                 logger.exception("Validation Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except (ValueError, InvokeError) as e:
-                if os.environ.get("DEBUG") and os.environ.get("DEBUG").lower() == "true":
+                if dify_config.DEBUG:
                     logger.exception("Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except Exception as e:
diff --git a/api/core/app/apps/completion/app_generator.py b/api/core/app/apps/completion/app_generator.py
index 46450d39c0..3bb05d05d8 100644
--- a/api/core/app/apps/completion/app_generator.py
+++ b/api/core/app/apps/completion/app_generator.py
@@ -1,5 +1,4 @@
 import logging
-import os
 import threading
 import uuid
 from collections.abc import Generator
@@ -8,6 +7,7 @@ from typing import Any, Literal, Union, overload
 from flask import Flask, current_app
 from pydantic import ValidationError
 
+from configs import dify_config
 from core.app.app_config.easy_ui_based_app.model_config.converter import ModelConfigConverter
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.base_app_queue_manager import AppQueueManager, GenerateTaskStoppedError, PublishFrom
@@ -203,7 +203,7 @@ class CompletionAppGenerator(MessageBasedAppGenerator):
                 logger.exception("Validation Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except (ValueError, InvokeError) as e:
-                if os.environ.get("DEBUG") and os.environ.get("DEBUG").lower() == "true":
+                if dify_config.DEBUG:
                     logger.exception("Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except Exception as e:
diff --git a/api/core/app/apps/workflow/app_generator.py b/api/core/app/apps/workflow/app_generator.py
index a865c8a68b..6e9c6804f9 100644
--- a/api/core/app/apps/workflow/app_generator.py
+++ b/api/core/app/apps/workflow/app_generator.py
@@ -1,6 +1,5 @@
 import contextvars
 import logging
-import os
 import threading
 import uuid
 from collections.abc import Generator, Mapping, Sequence
@@ -10,6 +9,7 @@ from flask import Flask, current_app
 from pydantic import ValidationError
 
 import contexts
+from configs import dify_config
 from core.app.app_config.features.file_upload.manager import FileUploadConfigManager
 from core.app.apps.base_app_generator import BaseAppGenerator
 from core.app.apps.base_app_queue_manager import AppQueueManager, GenerateTaskStoppedError, PublishFrom
@@ -261,7 +261,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
                 logger.exception("Validation Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except (ValueError, InvokeError) as e:
-                if os.environ.get("DEBUG") and os.environ.get("DEBUG", "false").lower() == "true":
+                if dify_config.DEBUG:
                     logger.exception("Error when generating")
                 queue_manager.publish_error(e, PublishFrom.APPLICATION_MANAGER)
             except Exception as e:
diff --git a/api/core/model_manager.py b/api/core/model_manager.py
index e21449ec24..059ba6c3d1 100644
--- a/api/core/model_manager.py
+++ b/api/core/model_manager.py
@@ -1,8 +1,8 @@
 import logging
-import os
 from collections.abc import Callable, Generator, Iterable, Sequence
 from typing import IO, Any, Optional, Union, cast
 
+from configs import dify_config
 from core.entities.embedding_type import EmbeddingInputType
 from core.entities.provider_configuration import ProviderConfiguration, ProviderModelBundle
 from core.entities.provider_entities import ModelLoadBalancingConfiguration
@@ -473,7 +473,7 @@ class LBModelManager:
 
                 continue
 
-            if bool(os.environ.get("DEBUG", "False").lower() == "true"):
+            if dify_config.DEBUG:
                 logger.info(
                     f"Model LB\nid: {config.id}\nname:{config.name}\n"
                     f"tenant_id: {self._tenant_id}\nprovider: {self._provider}\n"

From 0e8ab0588f0c2c7713b024164ece38be6f7df084 Mon Sep 17 00:00:00 2001
From: Bowen Liang <liangbowen@gf.com.cn>
Date: Fri, 8 Nov 2024 09:42:53 +0800
Subject: [PATCH 12/14] fix: (#10437 followup) fix conditions with DEBUG config
 (#10438)

---
 api/app.py         | 2 +-
 api/app_factory.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/api/app.py b/api/app.py
index 93774bd75d..60cd622ef4 100644
--- a/api/app.py
+++ b/api/app.py
@@ -2,7 +2,7 @@ import os
 
 from configs import dify_config
 
-if dify_config.DEBUG:
+if not dify_config.DEBUG:
     from gevent import monkey
 
     monkey.patch_all()
diff --git a/api/app_factory.py b/api/app_factory.py
index f53cf14f4b..60a584798b 100644
--- a/api/app_factory.py
+++ b/api/app_factory.py
@@ -2,7 +2,7 @@ import os
 
 from configs import dify_config
 
-if dify_config.DEBUG:
+if not dify_config.DEBUG:
     from gevent import monkey
 
     monkey.patch_all()

From c9f785e00f36258182c530ffd76d3dc4fa3ab52c Mon Sep 17 00:00:00 2001
From: "Leo.Wang" <wlrnet@163.com>
Date: Fri, 8 Nov 2024 09:53:03 +0800
Subject: [PATCH 13/14] Feat/tools/gitlab (#10407)

---
 api/core/rag/extractor/word_extractor.py      |  7 +-
 .../builtin/gitlab/tools/gitlab_commits.py    | 66 +++++++--------
 .../builtin/gitlab/tools/gitlab_commits.yaml  | 14 ++--
 .../gitlab/tools/gitlab_mergerequests.py      | 78 ++++++++++++++++++
 .../gitlab/tools/gitlab_mergerequests.yaml    | 77 ++++++++++++++++++
 .../builtin/gitlab/tools/gitlab_projects.py   | 81 +++++++++++++++++++
 .../builtin/gitlab/tools/gitlab_projects.yaml | 45 +++++++++++
 7 files changed, 320 insertions(+), 48 deletions(-)
 create mode 100644 api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.py
 create mode 100644 api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.yaml
 create mode 100644 api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.py
 create mode 100644 api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.yaml

diff --git a/api/core/rag/extractor/word_extractor.py b/api/core/rag/extractor/word_extractor.py
index 576bba0142..b59e7f94fd 100644
--- a/api/core/rag/extractor/word_extractor.py
+++ b/api/core/rag/extractor/word_extractor.py
@@ -28,7 +28,6 @@ logger = logging.getLogger(__name__)
 class WordExtractor(BaseExtractor):
     """Load docx files.
 
-
     Args:
         file_path: Path to the file to load.
     """
@@ -51,9 +50,9 @@ class WordExtractor(BaseExtractor):
 
             self.web_path = self.file_path
             # TODO: use a better way to handle the file
-            self.temp_file = tempfile.NamedTemporaryFile()  # noqa: SIM115
-            self.temp_file.write(r.content)
-            self.file_path = self.temp_file.name
+            with tempfile.NamedTemporaryFile(delete=False) as self.temp_file:
+                self.temp_file.write(r.content)
+                self.file_path = self.temp_file.name
         elif not os.path.isfile(self.file_path):
             raise ValueError(f"File path {self.file_path} is not a valid file or url")
 
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.py b/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.py
index 45ab15f437..716da7c8c1 100644
--- a/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.py
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.py
@@ -13,15 +13,15 @@ class GitlabCommitsTool(BuiltinTool):
     def _invoke(
         self, user_id: str, tool_parameters: dict[str, Any]
     ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
-        project = tool_parameters.get("project", "")
+        branch = tool_parameters.get("branch", "")
         repository = tool_parameters.get("repository", "")
         employee = tool_parameters.get("employee", "")
         start_time = tool_parameters.get("start_time", "")
         end_time = tool_parameters.get("end_time", "")
         change_type = tool_parameters.get("change_type", "all")
 
-        if not project and not repository:
-            return self.create_text_message("Either project or repository is required")
+        if not repository:
+            return self.create_text_message("Either repository is required")
 
         if not start_time:
             start_time = (datetime.utcnow() - timedelta(days=1)).isoformat()
@@ -37,14 +37,9 @@ class GitlabCommitsTool(BuiltinTool):
             site_url = "https://gitlab.com"
 
         # Get commit content
-        if repository:
-            result = self.fetch_commits(
-                site_url, access_token, repository, employee, start_time, end_time, change_type, is_repository=True
-            )
-        else:
-            result = self.fetch_commits(
-                site_url, access_token, project, employee, start_time, end_time, change_type, is_repository=False
-            )
+        result = self.fetch_commits(
+            site_url, access_token, repository, branch, employee, start_time, end_time, change_type, is_repository=True
+        )
 
         return [self.create_json_message(item) for item in result]
 
@@ -52,7 +47,8 @@ class GitlabCommitsTool(BuiltinTool):
         self,
         site_url: str,
         access_token: str,
-        identifier: str,
+        repository: str,
+        branch: str,
         employee: str,
         start_time: str,
         end_time: str,
@@ -64,27 +60,14 @@ class GitlabCommitsTool(BuiltinTool):
         results = []
 
         try:
-            if is_repository:
-                # URL encode the repository path
-                encoded_identifier = urllib.parse.quote(identifier, safe="")
-                commits_url = f"{domain}/api/v4/projects/{encoded_identifier}/repository/commits"
-            else:
-                # Get all projects
-                url = f"{domain}/api/v4/projects"
-                response = requests.get(url, headers=headers)
-                response.raise_for_status()
-                projects = response.json()
-
-                filtered_projects = [p for p in projects if identifier == "*" or p["name"] == identifier]
-
-                for project in filtered_projects:
-                    project_id = project["id"]
-                    project_name = project["name"]
-                    print(f"Project: {project_name}")
-
-                    commits_url = f"{domain}/api/v4/projects/{project_id}/repository/commits"
+            # URL encode the repository path
+            encoded_repository = urllib.parse.quote(repository, safe="")
+            commits_url = f"{domain}/api/v4/projects/{encoded_repository}/repository/commits"
 
+            # Fetch commits for the repository
             params = {"since": start_time, "until": end_time}
+            if branch:
+                params["ref_name"] = branch
             if employee:
                 params["author"] = employee
 
@@ -96,10 +79,7 @@ class GitlabCommitsTool(BuiltinTool):
                 commit_sha = commit["id"]
                 author_name = commit["author_name"]
 
-                if is_repository:
-                    diff_url = f"{domain}/api/v4/projects/{encoded_identifier}/repository/commits/{commit_sha}/diff"
-                else:
-                    diff_url = f"{domain}/api/v4/projects/{project_id}/repository/commits/{commit_sha}/diff"
+                diff_url = f"{domain}/api/v4/projects/{encoded_repository}/repository/commits/{commit_sha}/diff"
 
                 diff_response = requests.get(diff_url, headers=headers)
                 diff_response.raise_for_status()
@@ -120,7 +100,14 @@ class GitlabCommitsTool(BuiltinTool):
                                     if line.startswith("+") and not line.startswith("+++")
                                 ]
                             )
-                            results.append({"commit_sha": commit_sha, "author_name": author_name, "diff": final_code})
+                            results.append(
+                                {
+                                    "diff_url": diff_url,
+                                    "commit_sha": commit_sha,
+                                    "author_name": author_name,
+                                    "diff": final_code,
+                                }
+                            )
                     else:
                         if total_changes > 1:
                             final_code = "".join(
@@ -134,7 +121,12 @@ class GitlabCommitsTool(BuiltinTool):
                             )
                             final_code_escaped = json.dumps(final_code)[1:-1]  # Escape the final code
                             results.append(
-                                {"commit_sha": commit_sha, "author_name": author_name, "diff": final_code_escaped}
+                                {
+                                    "diff_url": diff_url,
+                                    "commit_sha": commit_sha,
+                                    "author_name": author_name,
+                                    "diff": final_code_escaped,
+                                }
                             )
         except requests.RequestException as e:
             print(f"Error fetching data from GitLab: {e}")
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.yaml b/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.yaml
index 669378ac97..2ff5fb570e 100644
--- a/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.yaml
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_commits.yaml
@@ -23,7 +23,7 @@ parameters:
     form: llm
   - name: repository
     type: string
-    required: false
+    required: true
     label:
       en_US: repository
       zh_Hans: 仓库路径
@@ -32,16 +32,16 @@ parameters:
       zh_Hans: 仓库路径，以namespace/project_name的形式。
     llm_description: Repository path for GitLab, like namespace/project_name.
     form: llm
-  - name: project
+  - name: branch
     type: string
     required: false
     label:
-      en_US: project
-      zh_Hans: 项目名
+      en_US: branch
+      zh_Hans: 分支名
     human_description:
-      en_US: project
-      zh_Hans: 项目名
-    llm_description: project for GitLab
+      en_US: branch
+      zh_Hans: 分支名
+    llm_description: branch for GitLab
     form: llm
   - name: start_time
     type: string
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.py b/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.py
new file mode 100644
index 0000000000..ef99fa82e9
--- /dev/null
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.py
@@ -0,0 +1,78 @@
+import urllib.parse
+from typing import Any, Union
+
+import requests
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class GitlabMergeRequestsTool(BuiltinTool):
+    def _invoke(
+        self, user_id: str, tool_parameters: dict[str, Any]
+    ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        repository = tool_parameters.get("repository", "")
+        branch = tool_parameters.get("branch", "")
+        start_time = tool_parameters.get("start_time", "")
+        end_time = tool_parameters.get("end_time", "")
+        state = tool_parameters.get("state", "opened")  # Default to "opened"
+
+        if not repository:
+            return self.create_text_message("Repository is required")
+
+        access_token = self.runtime.credentials.get("access_tokens")
+        site_url = self.runtime.credentials.get("site_url")
+
+        if not access_token:
+            return self.create_text_message("Gitlab API Access Tokens is required.")
+        if not site_url:
+            site_url = "https://gitlab.com"
+
+        # Get merge requests
+        result = self.get_merge_requests(site_url, access_token, repository, branch, start_time, end_time, state)
+
+        return [self.create_json_message(item) for item in result]
+
+    def get_merge_requests(
+        self, site_url: str, access_token: str, repository: str, branch: str, start_time: str, end_time: str, state: str
+    ) -> list[dict[str, Any]]:
+        domain = site_url
+        headers = {"PRIVATE-TOKEN": access_token}
+        results = []
+
+        try:
+            # URL encode the repository path
+            encoded_repository = urllib.parse.quote(repository, safe="")
+            merge_requests_url = f"{domain}/api/v4/projects/{encoded_repository}/merge_requests"
+            params = {"state": state}
+
+            # Add time filters if provided
+            if start_time:
+                params["created_after"] = start_time
+            if end_time:
+                params["created_before"] = end_time
+
+            response = requests.get(merge_requests_url, headers=headers, params=params)
+            response.raise_for_status()
+            merge_requests = response.json()
+
+            for mr in merge_requests:
+                # Filter by target branch
+                if branch and mr["target_branch"] != branch:
+                    continue
+
+                results.append(
+                    {
+                        "id": mr["id"],
+                        "title": mr["title"],
+                        "author": mr["author"]["name"],
+                        "web_url": mr["web_url"],
+                        "target_branch": mr["target_branch"],
+                        "created_at": mr["created_at"],
+                        "state": mr["state"],
+                    }
+                )
+        except requests.RequestException as e:
+            print(f"Error fetching merge requests from GitLab: {e}")
+
+        return results
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.yaml b/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.yaml
new file mode 100644
index 0000000000..4c886b69c0
--- /dev/null
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_mergerequests.yaml
@@ -0,0 +1,77 @@
+identity:
+  name: gitlab_mergerequests
+  author: Leo.Wang
+  label:
+    en_US: GitLab Merge Requests
+    zh_Hans: GitLab 合并请求查询
+description:
+  human:
+    en_US: A tool for query GitLab merge requests, Input should be a exists reposity or branch.
+    zh_Hans: 一个用于查询 GitLab 代码合并请求的工具，输入的内容应该是一个已存在的仓库名或者分支。
+  llm: A tool for query GitLab merge requests, Input should be a exists reposity or branch.
+parameters:
+  - name: repository
+    type: string
+    required: false
+    label:
+      en_US: repository
+      zh_Hans: 仓库路径
+    human_description:
+      en_US: repository
+      zh_Hans: 仓库路径，以namespace/project_name的形式。
+    llm_description: Repository path for GitLab, like namespace/project_name.
+    form: llm
+  - name: branch
+    type: string
+    required: false
+    label:
+      en_US: branch
+      zh_Hans: 分支名
+    human_description:
+      en_US: branch
+      zh_Hans: 分支名
+    llm_description: branch for GitLab
+    form: llm
+  - name: start_time
+    type: string
+    required: false
+    label:
+      en_US: start_time
+      zh_Hans: 开始时间
+    human_description:
+      en_US: start_time
+      zh_Hans: 开始时间
+    llm_description: Start time for GitLab
+    form: llm
+  - name: end_time
+    type: string
+    required: false
+    label:
+      en_US: end_time
+      zh_Hans: 结束时间
+    human_description:
+      en_US: end_time
+      zh_Hans: 结束时间
+    llm_description: End time for GitLab
+    form: llm
+  - name: state
+    type: select
+    required: false
+    options:
+      - value: opened
+        label:
+          en_US: opened
+          zh_Hans: 打开
+      - value: closed
+        label:
+          en_US: closed
+          zh_Hans: 关闭
+    default: opened
+    label:
+      en_US: state
+      zh_Hans: 变更状态
+    human_description:
+      en_US: state
+      zh_Hans: 变更状态
+    llm_description: Merge request state type for GitLab
+    form: llm
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.py b/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.py
new file mode 100644
index 0000000000..ea0c028b4f
--- /dev/null
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.py
@@ -0,0 +1,81 @@
+import urllib.parse
+from typing import Any, Union
+
+import requests
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class GitlabProjectsTool(BuiltinTool):
+    def _invoke(
+        self, user_id: str, tool_parameters: dict[str, Any]
+    ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        project_name = tool_parameters.get("project_name", "")
+        page = tool_parameters.get("page", 1)
+        page_size = tool_parameters.get("page_size", 20)
+
+        access_token = self.runtime.credentials.get("access_tokens")
+        site_url = self.runtime.credentials.get("site_url")
+
+        if not access_token:
+            return self.create_text_message("Gitlab API Access Tokens is required.")
+        if not site_url:
+            site_url = "https://gitlab.com"
+
+        # Get project content
+        result = self.fetch_projects(site_url, access_token, project_name, page, page_size)
+
+        return [self.create_json_message(item) for item in result]
+
+    def fetch_projects(
+        self,
+        site_url: str,
+        access_token: str,
+        project_name: str,
+        page: str,
+        page_size: str,
+    ) -> list[dict[str, Any]]:
+        domain = site_url
+        headers = {"PRIVATE-TOKEN": access_token}
+        results = []
+
+        try:
+            if project_name:
+                # URL encode the project name for the search query
+                encoded_project_name = urllib.parse.quote(project_name, safe="")
+                projects_url = (
+                    f"{domain}/api/v4/projects?search={encoded_project_name}&page={page}&per_page={page_size}"
+                )
+            else:
+                projects_url = f"{domain}/api/v4/projects?page={page}&per_page={page_size}"
+
+            response = requests.get(projects_url, headers=headers)
+            response.raise_for_status()
+            projects = response.json()
+
+            for project in projects:
+                # Filter projects by exact name match if necessary
+                if project_name and project["name"].lower() == project_name.lower():
+                    results.append(
+                        {
+                            "id": project["id"],
+                            "name": project["name"],
+                            "description": project.get("description", ""),
+                            "web_url": project["web_url"],
+                        }
+                    )
+                elif not project_name:
+                    # If no specific project name is provided, add all projects
+                    results.append(
+                        {
+                            "id": project["id"],
+                            "name": project["name"],
+                            "description": project.get("description", ""),
+                            "web_url": project["web_url"],
+                        }
+                    )
+        except requests.RequestException as e:
+            print(f"Error fetching data from GitLab: {e}")
+
+        return results
diff --git a/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.yaml b/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.yaml
new file mode 100644
index 0000000000..5fe098e1f7
--- /dev/null
+++ b/api/core/tools/provider/builtin/gitlab/tools/gitlab_projects.yaml
@@ -0,0 +1,45 @@
+identity:
+  name: gitlab_projects
+  author: Leo.Wang
+  label:
+    en_US: GitLab Projects
+    zh_Hans: GitLab 项目列表查询
+description:
+  human:
+    en_US: A tool for query GitLab projects, Input should be a project name.
+    zh_Hans: 一个用于查询 GitLab 项目列表的工具，输入的内容应该是一个项目名称。
+  llm: A tool for query GitLab projects, Input should be a project name.
+parameters:
+  - name: project_name
+    type: string
+    required: false
+    label:
+      en_US: project_name
+      zh_Hans: 项目名称
+    human_description:
+      en_US: project_name
+      zh_Hans: 项目名称
+    llm_description: Project name for GitLab
+    form: llm
+  - name: page
+    type: string
+    required: false
+    label:
+      en_US: page
+      zh_Hans: 页码
+    human_description:
+      en_US: page
+      zh_Hans: 页码
+    llm_description: Page index for GitLab
+    form: llm
+  - name: page_size
+    type: string
+    required: false
+    label:
+      en_US: page_size
+      zh_Hans: 每页数量
+    human_description:
+      en_US: page_size
+      zh_Hans: 每页数量
+    llm_description: Page size for GitLab
+    form: llm

From 033ab5490bf9b23516edbf1db0aaf7cf61721606 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E9=9D=9E=E6=B3=95=E6=93=8D=E4=BD=9C?= <hjlarry@163.com>
Date: Fri, 8 Nov 2024 13:22:52 +0800
Subject: [PATCH 14/14] feat: support LLM understand video (#9828)

---
 api/.env.example                              |  3 +-
 api/configs/feature/__init__.py               |  9 ++++-
 api/core/file/file_manager.py                 | 12 +++++-
 api/core/model_runtime/entities/__init__.py   |  2 +
 .../entities/message_entities.py              |  7 ++++
 .../model_providers/tongyi/llm/llm.py         |  9 +++++
 .../model_providers/zhipuai/llm/llm.py        | 38 +++++++++++++------
 api/core/workflow/nodes/llm/node.py           |  5 ++-
 .../components/app/configuration/index.tsx    |  4 +-
 9 files changed, 69 insertions(+), 20 deletions(-)

diff --git a/api/.env.example b/api/.env.example
index 1214154389..5dfc398df2 100644
--- a/api/.env.example
+++ b/api/.env.example
@@ -285,8 +285,9 @@ UPLOAD_IMAGE_FILE_SIZE_LIMIT=10
 UPLOAD_VIDEO_FILE_SIZE_LIMIT=100
 UPLOAD_AUDIO_FILE_SIZE_LIMIT=50
 
-# Model Configuration
+# Model configuration
 MULTIMODAL_SEND_IMAGE_FORMAT=base64
+MULTIMODAL_SEND_VIDEO_FORMAT=base64
 PROMPT_GENERATION_MAX_TOKENS=512
 CODE_GENERATION_MAX_TOKENS=1024
 
diff --git a/api/configs/feature/__init__.py b/api/configs/feature/__init__.py
index 5babae8810..f011b638e3 100644
--- a/api/configs/feature/__init__.py
+++ b/api/configs/feature/__init__.py
@@ -634,12 +634,17 @@ class IndexingConfig(BaseSettings):
     )
 
 
-class ImageFormatConfig(BaseSettings):
+class VisionFormatConfig(BaseSettings):
     MULTIMODAL_SEND_IMAGE_FORMAT: Literal["base64", "url"] = Field(
         description="Format for sending images in multimodal contexts ('base64' or 'url'), default is base64",
         default="base64",
     )
 
+    MULTIMODAL_SEND_VIDEO_FORMAT: Literal["base64", "url"] = Field(
+        description="Format for sending videos in multimodal contexts ('base64' or 'url'), default is base64",
+        default="base64",
+    )
+
 
 class CeleryBeatConfig(BaseSettings):
     CELERY_BEAT_SCHEDULER_TIME: int = Field(
@@ -742,7 +747,7 @@ class FeatureConfig(
     FileAccessConfig,
     FileUploadConfig,
     HttpConfig,
-    ImageFormatConfig,
+    VisionFormatConfig,
     InnerAPIConfig,
     IndexingConfig,
     LoggingConfig,
diff --git a/api/core/file/file_manager.py b/api/core/file/file_manager.py
index b69d7a74c0..ff9220d35f 100644
--- a/api/core/file/file_manager.py
+++ b/api/core/file/file_manager.py
@@ -3,7 +3,7 @@ import base64
 from configs import dify_config
 from core.file import file_repository
 from core.helper import ssrf_proxy
-from core.model_runtime.entities import AudioPromptMessageContent, ImagePromptMessageContent
+from core.model_runtime.entities import AudioPromptMessageContent, ImagePromptMessageContent, VideoPromptMessageContent
 from extensions.ext_database import db
 from extensions.ext_storage import storage
 
@@ -71,6 +71,12 @@ def to_prompt_message_content(f: File, /):
             if f.extension is None:
                 raise ValueError("Missing file extension")
             return AudioPromptMessageContent(data=encoded_string, format=f.extension.lstrip("."))
+        case FileType.VIDEO:
+            if dify_config.MULTIMODAL_SEND_VIDEO_FORMAT == "url":
+                data = _to_url(f)
+            else:
+                data = _to_base64_data_string(f)
+            return VideoPromptMessageContent(data=data, format=f.extension.lstrip("."))
         case _:
             raise ValueError(f"file type {f.type} is not supported")
 
@@ -112,7 +118,7 @@ def _download_file_content(path: str, /):
 def _get_encoded_string(f: File, /):
     match f.transfer_method:
         case FileTransferMethod.REMOTE_URL:
-            response = ssrf_proxy.get(f.remote_url)
+            response = ssrf_proxy.get(f.remote_url, follow_redirects=True)
             response.raise_for_status()
             content = response.content
             encoded_string = base64.b64encode(content).decode("utf-8")
@@ -140,6 +146,8 @@ def _file_to_encoded_string(f: File, /):
     match f.type:
         case FileType.IMAGE:
             return _to_base64_data_string(f)
+        case FileType.VIDEO:
+            return _to_base64_data_string(f)
         case FileType.AUDIO:
             return _get_encoded_string(f)
         case _:
diff --git a/api/core/model_runtime/entities/__init__.py b/api/core/model_runtime/entities/__init__.py
index b3eb4d4dfe..f5d4427e3e 100644
--- a/api/core/model_runtime/entities/__init__.py
+++ b/api/core/model_runtime/entities/__init__.py
@@ -12,11 +12,13 @@ from .message_entities import (
     TextPromptMessageContent,
     ToolPromptMessage,
     UserPromptMessage,
+    VideoPromptMessageContent,
 )
 from .model_entities import ModelPropertyKey
 
 __all__ = [
     "ImagePromptMessageContent",
+    "VideoPromptMessageContent",
     "PromptMessage",
     "PromptMessageRole",
     "LLMUsage",
diff --git a/api/core/model_runtime/entities/message_entities.py b/api/core/model_runtime/entities/message_entities.py
index cda1639661..3c244d368e 100644
--- a/api/core/model_runtime/entities/message_entities.py
+++ b/api/core/model_runtime/entities/message_entities.py
@@ -56,6 +56,7 @@ class PromptMessageContentType(Enum):
     TEXT = "text"
     IMAGE = "image"
     AUDIO = "audio"
+    VIDEO = "video"
 
 
 class PromptMessageContent(BaseModel):
@@ -75,6 +76,12 @@ class TextPromptMessageContent(PromptMessageContent):
     type: PromptMessageContentType = PromptMessageContentType.TEXT
 
 
+class VideoPromptMessageContent(PromptMessageContent):
+    type: PromptMessageContentType = PromptMessageContentType.VIDEO
+    data: str = Field(..., description="Base64 encoded video data")
+    format: str = Field(..., description="Video format")
+
+
 class AudioPromptMessageContent(PromptMessageContent):
     type: PromptMessageContentType = PromptMessageContentType.AUDIO
     data: str = Field(..., description="Base64 encoded audio data")
diff --git a/api/core/model_runtime/model_providers/tongyi/llm/llm.py b/api/core/model_runtime/model_providers/tongyi/llm/llm.py
index 3a1bb75a59..cde5d214d0 100644
--- a/api/core/model_runtime/model_providers/tongyi/llm/llm.py
+++ b/api/core/model_runtime/model_providers/tongyi/llm/llm.py
@@ -29,6 +29,7 @@ from core.model_runtime.entities.message_entities import (
     TextPromptMessageContent,
     ToolPromptMessage,
     UserPromptMessage,
+    VideoPromptMessageContent,
 )
 from core.model_runtime.entities.model_entities import (
     AIModelEntity,
@@ -431,6 +432,14 @@ class TongyiLargeLanguageModel(LargeLanguageModel):
 
                             sub_message_dict = {"image": image_url}
                             sub_messages.append(sub_message_dict)
+                        elif message_content.type == PromptMessageContentType.VIDEO:
+                            message_content = cast(VideoPromptMessageContent, message_content)
+                            video_url = message_content.data
+                            if message_content.data.startswith("data:"):
+                                raise InvokeError("not support base64, please set MULTIMODAL_SEND_VIDEO_FORMAT to url")
+
+                            sub_message_dict = {"video": video_url}
+                            sub_messages.append(sub_message_dict)
 
                     # resort sub_messages to ensure text is always at last
                     sub_messages = sorted(sub_messages, key=lambda x: "text" in x)
diff --git a/api/core/model_runtime/model_providers/zhipuai/llm/llm.py b/api/core/model_runtime/model_providers/zhipuai/llm/llm.py
index 43bffad2a0..eddb94aba3 100644
--- a/api/core/model_runtime/model_providers/zhipuai/llm/llm.py
+++ b/api/core/model_runtime/model_providers/zhipuai/llm/llm.py
@@ -313,21 +313,35 @@ class ZhipuAILargeLanguageModel(_CommonZhipuaiAI, LargeLanguageModel):
         return params
 
     def _construct_glm_4v_messages(self, prompt_message: Union[str, list[PromptMessageContent]]) -> list[dict]:
-        if isinstance(prompt_message, str):
+        if isinstance(prompt_message, list):
+            sub_messages = []
+            for item in prompt_message:
+                if item.type == PromptMessageContentType.IMAGE:
+                    sub_messages.append(
+                        {
+                            "type": "image_url",
+                            "image_url": {"url": self._remove_base64_header(item.data)},
+                        }
+                    )
+                elif item.type == PromptMessageContentType.VIDEO:
+                    sub_messages.append(
+                        {
+                            "type": "video_url",
+                            "video_url": {"url": self._remove_base64_header(item.data)},
+                        }
+                    )
+                else:
+                    sub_messages.append({"type": "text", "text": item.data})
+            return sub_messages
+        else:
             return [{"type": "text", "text": prompt_message}]
 
-        return [
-            {"type": "image_url", "image_url": {"url": self._remove_image_header(item.data)}}
-            if item.type == PromptMessageContentType.IMAGE
-            else {"type": "text", "text": item.data}
-            for item in prompt_message
-        ]
+    def _remove_base64_header(self, file_content: str) -> str:
+        if file_content.startswith("data:"):
+            data_split = file_content.split(";base64,")
+            return data_split[1]
 
-    def _remove_image_header(self, image: str) -> str:
-        if image.startswith("data:image"):
-            return image.split(",")[1]
-
-        return image
+        return file_content
 
     def _handle_generate_response(
         self,
diff --git a/api/core/workflow/nodes/llm/node.py b/api/core/workflow/nodes/llm/node.py
index 47b0e25d9c..eb4d1c9d87 100644
--- a/api/core/workflow/nodes/llm/node.py
+++ b/api/core/workflow/nodes/llm/node.py
@@ -14,6 +14,7 @@ from core.model_runtime.entities import (
     PromptMessage,
     PromptMessageContentType,
     TextPromptMessageContent,
+    VideoPromptMessageContent,
 )
 from core.model_runtime.entities.llm_entities import LLMResult, LLMUsage
 from core.model_runtime.entities.model_entities import ModelType
@@ -560,7 +561,9 @@ class LLMNode(BaseNode[LLMNodeData]):
                         # cuz vision detail is related to the configuration from FileUpload feature.
                         content_item.detail = vision_detail
                         prompt_message_content.append(content_item)
-                    elif isinstance(content_item, TextPromptMessageContent | AudioPromptMessageContent):
+                    elif isinstance(
+                        content_item, TextPromptMessageContent | AudioPromptMessageContent | VideoPromptMessageContent
+                    ):
                         prompt_message_content.append(content_item)
 
                 if len(prompt_message_content) > 1:
diff --git a/web/app/components/app/configuration/index.tsx b/web/app/components/app/configuration/index.tsx
index 639cb2fad1..2bb11a870c 100644
--- a/web/app/components/app/configuration/index.tsx
+++ b/web/app/components/app/configuration/index.tsx
@@ -468,8 +468,8 @@ const Configuration: FC = () => {
           transfer_methods: modelConfig.file_upload?.image?.transfer_methods || ['local_file', 'remote_url'],
         },
         enabled: !!(modelConfig.file_upload?.enabled || modelConfig.file_upload?.image?.enabled),
-        allowed_file_types: modelConfig.file_upload?.allowed_file_types || [SupportUploadFileTypes.image],
-        allowed_file_extensions: modelConfig.file_upload?.allowed_file_extensions || FILE_EXTS[SupportUploadFileTypes.image].map(ext => `.${ext}`),
+        allowed_file_types: modelConfig.file_upload?.allowed_file_types || [SupportUploadFileTypes.image, SupportUploadFileTypes.video],
+        allowed_file_extensions: modelConfig.file_upload?.allowed_file_extensions || [...FILE_EXTS[SupportUploadFileTypes.image], ...FILE_EXTS[SupportUploadFileTypes.video]].map(ext => `.${ext}`),
         allowed_file_upload_methods: modelConfig.file_upload?.allowed_file_upload_methods || modelConfig.file_upload?.image?.transfer_methods || ['local_file', 'remote_url'],
         number_limits: modelConfig.file_upload?.number_limits || modelConfig.file_upload?.image?.number_limits || 3,
         fileUploadConfig: fileUploadConfigResponse,