AI Assistant Service overview,DataWorks AI Ops Assistant,OpenClaw - DataWorks

AI Assistant Service is an AI-powered assistant for DataWorks, built on the OpenClaw framework. The service integrates with instant messaging (IM) platforms such as DingTalk and Feishu, letting you use natural language to perform tasks like task diagnostics, alert analysis, and operations and maintenance (O&M). This shifts your workflow from reactive alert response to proactive, intelligent operations.

Features

AI Assistant Service is a dedicated intelligent assistant launched after integrating OpenClaw with DataWorks. Unlike the built-in DataWorks Agent in the DataWorks console, AI Assistant Service extends AIOps capabilities to IM platforms like DingTalk and Feishu. This allows you to perform O&M tasks without logging in to the DataWorks console.

Core benefits

Instant response: Receive alerts and run AI-powered diagnostics directly within your IM group chat, reducing alert response times from minutes to seconds.
Intelligent diagnostics: Automatically analyze the root cause of task failures and the impact of alerts, and receive recommendations for remediation.
Interactive O&M: Use natural language commands for O&M tasks, such as rerunning tasks or modifying resource groups. All write operations require secondary confirmation.
Data security: All data is processed within the DataWorks platform. AI inference runs on the platform's internal network, and the service desensitizes sensitive data before sending it to the AI.

Available regions and policies

Important

AI Assistant Service is currently in a phased release. If the service is not available in your region, you can click the application link to apply for a trial, or join the AI Assistant Service DingTalk group for more technical support.

Eligible users: Customers with DataWorks Basic Edition or higher.
Available regions: China (Zhangjiakou), China (Beijing), China (Ulanqab), China (Hangzhou), China (Shanghai), China (Shenzhen), China (Chengdu), China (Hong Kong), Singapore, Malaysia (Kuala Lumpur), Indonesia (Jakarta), and Japan (Tokyo).

Core capabilities

AI Assistant Service supports the following eight core scenarios:

Scenario	Core capability	Example command
Fault diagnostics (core strength)	Alert diagnostics (root cause and upstream analysis), instance diagnostics (single and batch), and project health diagnostics.	`Diagnose the failure reason for instance 12345678`
Task scheduling	Query or modify task configurations; query, rerun, terminate, or set instances to success; freeze or unfreeze tasks; adjust priority; view dependency topology; and migrate resource groups.	`Rerun instance 12345678`
Workflow management	Query or modify workflows, backfill data, perform smoke tests, manually trigger, start, stop, or rerun workflows, and refresh dependencies.	`Backfill data for workflow wf-001 from 2026-01-01 to 2026-01-31`
Monitoring and O&M	Query baseline configurations and Service Level Agreements (SLAs), manage alert rules, query alert messages, and query on-call schedules.	`Check the SLA status for baseline base_test`
Data quality	Manage data quality monitoring rules, manage quality alert rules, and automatically configure quality monitoring.	`Query DQC rule configuration for table ods_user_info`
Data governance	Manage labels (create, bind, unbind), query governance issues and rules, analyze assets (storage, compute, sync, scheduling), and view asset health scores.	`View the data governance score for the project`
Project management	Query workspace lists and details, query member and role lists, and query current user information.	`Show member list for workspace xxx`
Resource management	Query resource group lists and details, and monitor resource group usage levels.	`Query the load status of resource group rg-001`

Note

The following capabilities of AI Assistant Service are currently read-only: member permission management and role management in project management, and network configuration in resource management. AI Assistant Service does not currently support creating workspaces, creating new tasks or nodes, or deleting workspaces or tasks.

Supported IM platforms

IM platform	Required configuration	Configuration reference
DingTalk	Client ID, Client Secret	Integrate AI Assistant Service with DingTalk
Feishu/Lark	App ID, App Secret	Integrate AI Assistant Service with Feishu
WeCom	Bot ID, Secret	Integrate AI Assistant Service with WeCom

Configuration

Prerequisites

Ensure you have a DataWorks Data Agent seat. If you do not have a seat, see Data Agent pricing.
Prepare a Serverless resource group to run the AI Assistant Service instance. The instance requires a minimum of 2 CUs.
You have created a bot application in DingTalk or Feishu and obtained the key configuration information, such as the Client ID and Client Secret for DingTalk or the App ID and App Secret for Feishu.

Create an instance

Important

Only primary accounts and RAM users with the AliyunDataWorksFullAccess policy can create and manage AI Assistant Service instances.

Log in to the DataWorks console, select the target region, and in the left-side navigation pane, click AI Assistant Service.
On the AI Assistant Service page, click Create Instance.

In the Create AI Assistant Service Instance dialog box, configure the following parameters:

Parameter	Description
Instance Name	Required. The name of the AI Assistant Service instance, used to identify and differentiate instances.
Resource Group	Required. Select the exclusive resource group to run the AI Assistant Service instance. The instance requires a minimum of 2 CUs. You can adjust the CU allocation from 2 CUs up to the maximum CUs of the resource group. You can view the resource usage of the AI Assistant Service in the resource group details. Note If the selected resource group does not have enough available CUs to meet the resource requirements of the instance, the resource group will not be selectable.
IM Channel	Required. Select the IM channel for the service. Currently, DingTalk, Feishu, Lark, and WeCom are supported. You can configure multiple IM channels simultaneously.
IM Configuration	Required. Enter the information for your selected IM channel: DingTalk: Enter the Client ID and Client Secret. Feishu/Lark: Enter the App ID and App Secret. WeCom: Enter the Bot ID and Secret. After configuration, you can click Test Connection to verify network connectivity. If your resource group does not have public network access, see Configure a Virtual Private Cloud (VPC) to access IM endpoints to allow VPC resources to access IM platform endpoints over a private network.
Execution Identity	Specify the identity used to access DataWorks. By default, this is the currently logged-in user. To define the access scope more precisely, use a RAM role. When creating the role, you can specify the workspaces and workspace roles it can access. AI Assistant Service uses this identity for permission checks and to execute operations in DataWorks.

Click OK.
Creating the instance takes about 1-3 minutes. Once complete, you can start interacting with the bot.

Manage AI Assistant Service instances

On the AI Assistant Service page, you can view and manage your created instances.

Start/Stop an instance: An instance starts by default after creation. If you modify the bot or instance configuration, you can stop and restart the instance for the changes to take effect.
Manage Execution Role: After you create an instance, if you need to modify the access permissions of the specified identity, click Manage Execution Role in the Actions column. You can edit the list of accessible workspaces and workspace roles.
Edit instance: Click Edit in the Actions column of the target instance to modify its name and description.
Delete instance: Click Delete in the Actions column of the target instance to delete it.

Important

Only the instance owner, the primary account, or users with AliyunDataWorksFullAccess permissions can edit or delete an instance.

Using AI Assistant Service via IM

After you create an AI Assistant Service instance, you can interact with it using natural language in a group chat where the bot is configured or in a private chat with the bot.

Ask questions proactively: In a group chat, @mention the bot or start a private chat and enter your question or command. AI Assistant Service understands your intent and executes the corresponding operation.

Scenario	Example command
Task failure diagnosis	`Diagnose the failure reason for instance 12345678`
Alert analysis	`Analyze current unresolved alerts`
Baseline SLA query	`Check the SLA status for baseline base_test`
Workflow data backfill	`Backfill data for workflow wf-001 from 2026-01-01 to 2026-01-31`
Rerun failed tasks	`Rerun today's failed scheduling instances in the abc workspace`

Automated alert diagnostics: When a DataWorks monitoring rule triggers an alert, AI Assistant Service automatically receives the alert, performs an AI-powered diagnosis, and pushes the result to the IM group.
Tell AI Assistant Service: Enable automatic monitoring alerts and push them to this conversation.
AI Assistant Service automatically configures an alert listener and sends alerts triggered by DataWorks to the current conversation as message cards. You can then use natural language commands in the chat (such as "view failure log," "rerun current node," or "roll back to previous version") to call the underlying APIs and perform O&M actions. This eliminates the need to log in to the console and shortens the time to resolve failures.

Important

Secondary confirmation for write operations: Before executing any write operation (such as rerun, freeze, delete, or modify), AI Assistant Service clearly displays the operation type, target object, and the number of items affected. The operation proceeds only after you provide explicit confirmation.

Billing

Fees for AI Assistant Service consist of the following components:

Billable item	Billing method	Description
Data Agent seat	Charged per seat	A Data Agent seat is a prerequisite for using AI Assistant Service. For more information, see Data Agent overview.
Resource group fees	pay-as-you-go (by duration)	Covers the CUs consumed by the AI Assistant Service instance in the resource group.
AI inference	Charged per token usage	Fees for calling AI models for intelligent diagnostics and analysis. For more information, see Data Agent pricing.

AI Assistant vs. DataWorks Agent

Comparison	DataWorks Agent	AI Assistant Service
Entry point	Within the DataWorks console (DataStudio page)	IM platforms (DingTalk, Feishu group chats)
Primary scenarios	All scenarios, including data development, data integration, and data governance.	O&M scenarios (alert diagnostics, task management, baseline monitoring) and data governance scenarios.
Interaction method	In-browser Copilot chat window	@mentioning the assistant in an IM group chat or receiving automated alerts.
Alert integration	Not supported	Supports receiving alerts and performing automated AI diagnostics.

AI Assistant Service DingTalk group

You must first click the application link to join the "Alibaba Cloud Big Data and AI Platform" organization. Then, scan the QR code below to join the AI Assistant Service DingTalk group for exclusive technical support.