The AI assistant service is an AI assistant built by DataWorks on the OpenClaw framework. You can chat with it in the DataWorks console or connect it to IM platforms such as DingTalk and Feishu. From an IM client, you can use natural language to analyze alerts, diagnose tasks, modify code, manage tasks, and govern data. DataWorks provides one-click network connectivity between resource groups and IM clients, knowledge base and Skill management, scheduled task management, and operation log auditing to help you build a dedicated enterprise AI assistant.
Features
The AI assistant service is a dedicated assistant built by integrating OpenClaw with DataWorks. Unlike the Agent built into the DataWorks console, it extends DataWorks capabilities to the web client and IM clients (DingTalk, Feishu, and WeCom). You can complete tasks without working in the DataWorks console.
Core values
Access from multiple clients: Chat with the AI assistant in the DataWorks console or in DingTalk, Feishu, or WeCom group chats.
Instant response: Receive alerts and run AI diagnoses directly in an IM group, reducing alert response time from minutes to seconds.
Intelligent diagnosis: Automatically analyze the root causes of task failures and the scope of an alert's impact, and suggest fixes.
Interactive O&M: Issue O&M instructions in natural language, such as rerunning a task or changing a resource group. Every write operation requires a second confirmation.
Knowledge enhancement: Automatically build a knowledge base (RAG) and manage memory to help the AI assistant understand your business context.
Scheduled inspections: Use built-in scheduled and triggered task templates for SLA monitoring, asset governance, cost analysis, and other scenarios, and receive scheduled inspection reports.
Data security: All data flows within DataWorks. AI inference runs on the platform's internal network, and sensitive data is masked before it is sent to AI.
Available regions and policies
Join the AI assistant service DingTalk group for more technical support.
Available to customers using 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), Japan (Tokyo), and US (Virginia).
Core capabilities
The data O&M Agent in the AI assistant service supports the following scenarios:
Scenario | Core capabilities | Example command |
Fault diagnosis (core strength) | Alert diagnosis (root cause identification and upstream analysis), instance diagnosis (single or batch), and workspace health diagnosis |
|
Task scheduling | Query or modify task configurations; query, rerun, terminate, or mark instances as succeeded; freeze or unfreeze tasks; adjust priorities; view dependency topologies; and migrate resource groups |
|
Workflow management | Query or modify workflows; backfill data or run smoke tests; manually trigger, start, stop, or rerun workflows; and refresh dependencies |
|
Monitoring and O&M | Query baseline configurations and SLAs; manage alert rules; query alert messages, duty rosters, and schedules |
|
Data quality | Manage data quality monitoring rules and quality alert rules, and automatically configure quality monitoring |
|
Data governance | Create, bind, or unbind tags; query governance issues and rules; analyze assets by storage, compute, synchronization, and scheduling; view asset health scores; and run governance inspections |
|
Project management | Query workspace lists and details, member and role lists, and current user information |
|
Resource management | Query resource group lists and details, and monitor resource group utilization |
|
The AI assistant service currently provides read-only access to member permission and role management under project management, and network configuration under resource management. It does not yet support creating workspaces or new tasks or nodes, or deleting workspaces or tasks.
Supported IM platforms
IM platform | Required configuration | Configuration reference |
DingTalk | Client ID, Client Secret | |
Feishu/Lark | App ID, App Secret | |
WeCom | Bot ID, Secret |
Prerequisites
Make sure you have enabled DataWorks Data Agent. If you have not enabled it, see Data Agent fees.
Prepare a serverless resource group to run an AI assistant service instance. The instance requires at least 2 CUs of resources.
To use the service through an IM client, first create a bot application in DingTalk or Feishu and obtain its credentials, such as the AppKey and AppSecret for DingTalk or the App ID and App Secret for Feishu.
Create an instance
Log on to the DataWorks console, select the destination region, and click AI assistant service.
On the AI assistant service page, click Create.
In the Create AI assistant service instance dialog box, configure the following settings:
Parameter
Description
Instance Name
Required. Enter a name to identify the AI assistant service instance. The name must start with a letter or Chinese character and can contain letters, Chinese characters, digits, underscores (_), and hyphens (-). It can be up to 128 characters long.
Resource Group
Required. Select the serverless resource group on which to run the AI assistant service instance. The instance requires at least 2 CUs.
NoteIf the selected resource group does not have enough available CUs for the instance, you cannot select that group.
Execution Identity
Required. Specify the identity that the AI assistant service uses to access DataWorks. For a personal assistant, we recommend Current Logon User. To share the assistant with multiple users, we recommend Create New RAM Role. You can specify the workspaces and workspace roles that this RAM role can access.
Instance description
Optional. The description of the instance. The description can be up to 256 characters in length.
Click OK.
Creating an instance takes approximately 1 to 3 minutes. You can use the AI assistant after creation is complete.
Manage instances
Basic operations
On the AI assistant service page, you can view and manage the instances that you created.
Start/Stop an instance: An instance starts by default after creation. After changing its configuration, stop and restart it for the changes to take effect.
Upgrade an instance: When a newer instance image is available, click the Upgradable label next to the Image column to upgrade the instance and use new features.
Execution Role Management: Change the access permissions of the specified execution identity, including its accessible workspaces and workspace roles.
Edit instance: Modify the instance name and description.
Delete instance: Click Delete in the Actions column of the target instance to delete the instance.
Retry: For instances that failed to be created or started, you can click Retry to re-execute the operation.
Channel configuration
You can configure an IM channel for the AI assistant service and chat with the assistant directly in DingTalk, Feishu/Lark, or WeCom. This lets you perform O&M work without using the DataWorks console. You can still use the assistant in the web client without an IM channel.
Click Configure for any channel and enter its required IM credentials in the dialog box:
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, click Test Network Connectivity to test the connection. If the network is not connected, click One-Click Network Setup to configure connectivity using one of the following methods:
PrivateZone + PrivateLink (recommended): If the resource group is in a VPC without public network access, use this method to let resources in the VPC access IM platform endpoints entirely over a private network, so traffic does not traverse the public Internet.
This method requires a vSwitch in the specified zone. If you have not created one, follow the on-screen instructions to create it.
Enable public network access: Use a NAT gateway and an EIP to enable public network access. This is easier to configure but less secure.
If the AI assistant service runs inspection or scheduled tasks and needs to @mention the responsible person in the relevant chat group, configure account mapping. For instructions, see Configure IM account mapping for @-mentions in AI Assistant.
Different network connectivity methods have different charges. For details, see Billing information.
Skill management
A Skill is a capability extension that the AI assistant can invoke. Skill management lets you extend the assistant for data O&M and governance. The AI assistant service provides built-in system Skills and supports custom Skills. The platform provides and maintains system Skills for scenarios such as data development assistance and product Q&A. You can upload custom capability modules to meet your business needs.
Click an instance name to open its details page, and then click the Skill management tab.
The Skill management tab is available only when the instance image is version v3.0.0 or later. If the tab is absent, upgrade the instance image first.
System skills
DataWorks provides and maintains the following built-in system Skills:
Data Development Skill: Helps with data development tasks such as code analysis and SQL generation.
Full-Series Product QA Skill: Answers questions about using DataWorks products.
You cannot delete system Skills, but you can enable or disable them with the toggle.
Custom skills
Import a Skill
In the Skill management tab, click Import a Skill.
In the Import Skill dialog box, configure the following parameters:
Parameter
Type
Required
Description
Name
Text field
Yes
A name that identifies the Skill. Up to 64 characters.
Description
Text area
No
A description of the Skill and its purpose. Up to 256 characters.
Skill file
File upload
Yes
Upload a compressed Skill package in .zip format. Drag-and-drop upload is supported.
After the configuration is complete, click Confirm to complete the import.
Enable/Disable a Skill
In the Skill list, use the toggle on a Skill card to enable or disable it. Once disabled, the AI assistant no longer invokes that Skill.
Upgrade a Skill
When a new version of a system Skill is available, the Skill card displays an Upgradable tag is displayed on the Skill card.
Click Upgrade button.
The system automatically disables and re-enables the Skill. The upgraded Skill then takes effect.
Delete a custom Skill
In the Skill list, find the custom Skill that you want to delete, click the Delete button, and confirm the deletion.
System Skills cannot be deleted. Only custom Skills can be deleted.
Get more Skills
In the Skill management tab, click Alibaba Cloud Skill Portal to go to the Alibaba Cloud Skill Portal page and browse or obtain more available Skill modules.
Memory management
Long-term memory stores important business context for an AI assistant instance. In memory management, you can save project assets, data standards, business processes, and other knowledge as files. The assistant can use this context across sessions to provide more accurate and personalized O&M recommendations.
On the AI assistant service list page, click an instance name to open its details page, and then click the Memory management tab.
Browse directory: Browse memory files in a tree directory. You can open subdirectories and return to the parent directory.
View file: Click a file to view its contents. Large files are truncated and display a truncation notice.
Import memory: Upload a file to a specified path. You can export memory from a local OpenClaw installation or another AI assistant service and import it for reuse.
Export: Download a single file or export multiple files as a package.
Delete: Delete one file or multiple files.
Knowledge base management
Knowledge bases have been separated from the AI assistant service into an independent module, Context Graph (Knowledge Base), which is used to create and manage them. You can use Share to authorize an AI assistant service or Data Agent to use a knowledge base created in Context Graph (Knowledge Base). For features, creation, and usage instructions, see Create a knowledge base.
New AI assistant services no longer create a local knowledge base. To provide knowledge to an assistant, create a knowledge base in Context Graph (Knowledge Base) and use Share to authorize the relevant AI assistant service instance.
For existing AI assistant service users, knowledge bases created previously within the service remain available as local knowledge bases on the Knowledge base tab of the instance details page. Existing use is unaffected.
Existing local knowledge bases support browsing, editing, regeneration, and clearing.
Instance monitoring
On the AI assistant service list page, click an instance name to open its details page, and then click the Monitoring tab to open instance monitoring and view resource usage:
CPU usage: Used CPU / total specification capacity.
Memory usage: Used memory / total specification capacity.
Disk usage: Used disk space / total specification capacity.
Select a time range to view historical trends. By default, data from the most recent hour is displayed.
Instance logs
On the AI assistant service list page, click an instance name to open its details page, and then click the Logs tab to enter instance monitoring and view resource usage. You can also query instance runtime logs:
You can filter log sources by subsystem.
You can search log content by keyword.
You can select a time range to filter logs.
Use the AI assistant service
Use conversations
Use the AI Assistant service from the Web client
After the AI Assistant service instance is created, click Conversation in the instance list to start a conversation with the AI assistant directly in the DataWorks console.
Session management
The web client manages regular conversations and scheduled tasks. The conversation sidebar provides the following features:
Regular sessions:
Create a session: Click + New Task to create a new conversation session.
Switch sessions: Click a session name in the session list to switch to the corresponding session.
Scheduled tasks:
Scheduled/triggered tasks: DataWorks provides templates for scheduled tasks. Select a template to start a new task quickly.
Conversation interaction
In the conversation interface, you can interact with the AI assistant in natural language. It invokes DataWorks APIs to perform O&M operations and asks for your confirmation before every write operation.
Use the AI Assistant service from IM clients
After creating an AI assistant service instance, you can interact with it in natural language in a group chat that has the bot configured or in a private chat with the bot.
Proactive inquiry: @mention the bot in a group or send it a private message with a question or instruction. The AI assistant service interprets your request and performs the corresponding operation.
Scenario
Example command
Task failure diagnosis
Diagnose why task instance 12345678 failedAlert analysis
Analyze currently unresolved alertsBaseline SLA query
Check whether baseline base_test meets its SLAWorkflow backfill
Backfill workflow wf-001 for the time range 2026-01-01 to 2026-01-31Rerun failed tasks
Rerun the failed scheduled instances in workspace abc todayAutomatic alert diagnosis:When a DataWorks monitoring rule triggers an alert, the AI assistant service receives it, runs an AI diagnosis, and sends the result to an IM group.
Tell the AI assistant service:
Enable automatic monitoring alerts and push them to the current conversation.The 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 instructions in the chat, such as "view the failure logs," "rerun the current node," or "roll back to the previous version," to invoke the underlying APIs for O&M without logging on to the console and reduce recovery time.
Enable governance inspections:When automatic inspections are enabled, the AI assistant service runs asset governance inspections and sends the results to an IM group on a schedule.
Tell the AI assistant service: Enable automatic inspections and push results to the current conversation.
By default, governance inspection results are sent to the IM client at 09:30 every day. You can change the time and frequency through conversation. You can also request a more specific governance check, such as "Enable governance inspections and report the number of tables owned by the primary account every three hours."
Write operation confirmation: Before any write operation, such as rerunning, freezing, deleting, or modifying, the AI assistant service shows the operation type, targets, and number affected. It executes the operation only after you explicitly confirm it.
Scheduled tasks
The AI assistant service provides task templates for quickly setting up scheduled inspections. In the conversation sidebar of the web client, click Scheduled tasks to open the template panel. Select a template to create a scheduled task conversation.
To configure a scheduled task in an IM client, instruct the AI assistant in natural language.
Scheduled inspection task templates
By default, the following task templates run every day at 09:00, generate inspection reports, and send them to the conversation:
Task name | Description |
SLA baseline completion report | Assess the overall completion of baselines in the workspace for today and generate a report |
Workspace completion report | Analyze the completion of today's workspace tasks and generate a report |
Inspect unhealthy assets | Summarize governance issues under your account, prioritize them, and generate a report |
Inspect data quality | Identify high-priority tables without data quality monitoring, analyze them, and recommend actions |
Daily root-cause analysis of failed tasks | Summarize yesterday's failed instances and analyze the five tasks that fail most frequently |
Daily resource group utilization inspection | Identify resource group utilization, load, and queue bottlenecks, and recommend scaling or scheduling changes |
Daily ranking of slow tasks | Compare historical run times and analyze the ten tasks with the largest unusual increases in run and wait times |
Daily inspection of alert coverage gaps | Find alert coverage gaps for key baselines and high-priority tasks, and recommend how to close them |
Daily inspection of backfill status | Summarize the status of backfill tasks over the past seven days and analyze failures |
Daily inspection of tables without lifecycle settings | Count tables without lifecycle settings by storage usage and list the top 20 with configuration recommendations |
Daily inspection of ownerless tables and obsolete assets | Identify tables that have not been accessed for a long time or have no owner, and prepare a list of candidates for decommissioning |
Daily audit of code changes against governance rules | Check governance rule results for node changes over the past 24 hours and summarize them by owner |
Daily ranking of storage and compute consumption | List the top 20 storage and compute consumers, analyze growth trends, and suggest optimizations |
Daily inspection of asset catalog coverage | Count uncataloged tables and their percentage, and recommend cataloging actions |
Custom scheduled tasks
DataWorks also supports custom scheduled tasks. In the web client, click Scheduled tasks, and then click Create custom task in the upper-right corner. Follow the AI assistant's prompts to enter the run interval and task details.
Billing information
The AI assistant service incurs the following charges:
Billable item | Billing method | Description |
Data Agent seat | Purchased by seat | A Data Agent seat is required before you can use the AI assistant service. For details, see Data Agent overview. |
Resource group fees | Pay-as-you-go (by duration) | Charges for the resource group CUs used while an AI assistant service instance runs. Connecting the resource group to a network may incur additional charges: PrivateZone + PrivateLink: PrivateZone (Billing), and endpoints (PrivateLink billing). Public network access: NAT (NAT Gateway billing). |
AI inference | Billed by token usage | Charges for AI model calls made by the AI assistant service for intelligent diagnosis and analysis. For details, see Data Agent fees. |
Differences from DataWorks Agent
Comparison item | DataWorks Agent | AI assistant service |
Access entry | DataWorks console (Data Studio page) | Web client (AI assistant service page in the DataWorks console) and IM clients (DingTalk, Feishu, and WeCom group chats) |
Primary scenarios | Data development, data integration, data governance, and other scenarios | O&M scenarios such as alert diagnosis, task management, and baseline monitoring, plus data governance |
Interaction method | Data Agent chat box in the browser | Dedicated web chat interface or @mention the AI assistant in an IM group |
Alert integration | Not supported | Receives alerts and runs automatic AI diagnoses |
Knowledge base | Not supported | Can build a workspace-level knowledge base (RAG) |
Scheduled tasks | Not supported | Supports scheduled inspections and triggered tasks |
Custom skills | Not supported | Supports importing and managing custom Skills |
AI assistant service DingTalk group
First click the application link to join the "Alibaba Cloud Big Data AI Platform" organization. Then scan the QR code below to join the AI assistant service DingTalk group and receive product support.
