DataBridge Agent is a multi-source data ingestion and parsing tool provided by AI-Native Database Service. It integrates core capabilities such as webpage data analysis, complex file parsing, and DTS task management. This topic describes how to create a DataBridge Agent instance and use its core features.
Prerequisites
-
An Alibaba Cloud account is created and real-name verification is complete.
-
(Optional) To manage Data Transmission Service (DTS) tasks through DataBridge Agent, prepare an AccessKey pair for a RAM user that has DTS management permissions.
Limits
During the invitation-only testing period, each Alibaba Cloud account can create up to 10 DataBridge Agent instances.
Procedure
Step 1: Create a DataBridge Agent instance
-
Log on to the AI-Native Database Service console.
-
In the left-side navigation pane, in the Data Agent section, click Data Ingestion.
-
On the overview page, click Create Instance on the right. Enter an instance name and select a region, and then click Create.
NoteInstance initialization typically takes 5 to 10 minutes.
Step 2: Start using the agent
After the DataBridge Agent instance is created, click the instance name or click New Session on the right to open the conversation panel. You can enter commands in the conversation panel to use the core features of DataBridge Agent.
(Optional) Step 3: Configure access credentials
To manage DTS tasks through DataBridge Agent, you must configure the AccessKey pair of a RAM user in the instance.
-
If you only need to query configuration or monitoring information for DTS tasks, you can grant read-only permissions for DTS (
AliyunDTSReadOnlyAccess). -
If you grant management permissions for DTS (
AliyunDTSFullAccess) to a RAM user, DataBridge Agent can perform high-risk operations such as modifying task parameters, stopping tasks, or releasing tasks based on your commands. Proceed with caution.
Run the following commands in the conversation panel to complete the configuration:
-
Run the following command to start the configuration:
aliyun configure -
Enter the AccessKey ID, AccessKey secret, and default region (Region ID) of the RAM user as prompted:
AccessKey ID xxx AccessKey Secret xxx Region ID xxx
Core features
DataBridge Agent provides three types of core features. You can use these features by entering natural language commands in the conversation panel.
Webpage data analysis
Enter a webpage URL and your analysis requirements. DataBridge Agent automatically collects the webpage content and generates an analytical report.
Example
Analyze the data on https://www.alibabacloud.com/help/zh/dts/product-overview/what-is-dts. What features does Alibaba Cloud DTS support?
Complex file parsing
DataBridge Agent includes a dedicated DTS skill (aliyun-dts-file-parser) to parse PDFs and other documents into Markdown for document content extraction and analysis.
Example
Parse file https://xxx.oss-cn-beijing.aliyuncs.com/example.pdf
DTS task management
DataBridge Agent includes a dedicated DTS skill (aliyun-dts-task-manager) for querying and managing your DTS migration, synchronization, and change tracking tasks.
-
Before you use this feature, complete Step 3 to configure access credentials.
-
When you manage a DTS task, enter the DTS instance ID, not the task name.
-
You can manage only tasks for which the precheck has passed and a DTS instance has been purchased.
Example
Query information for DTS task dtsboss6a15****
Appendix: Dedicated Skill list
|
Skill name |
Description |
|
|
Parses PDFs and other documents into Markdown. Used for document content extraction, PDF-to-text conversion, and document analysis. |
|
|
Manages DTS migration, synchronization, and change tracking tasks. Supports task queries, status monitoring, task suspension, and task release. |
|
|
Automatically collects webpage content based on a URL. Uses a progressive intelligent extraction strategy that automatically identifies and switches to the optimal parsing method. |