All Products
Search

This Product

    Data Management:Meta Agent

    Document Center

    Data Management:Meta Agent

    Last Updated:Feb 04, 2026

    DMS Meta Agent is an enterprise-grade data management agent powered by a large language model (LLM). Using automated data asset inventory and natural language interaction, Meta Agent transforms complex data assets into business knowledge that is easy to understand and use. It allows users to find, understand, and use data securely and efficiently, similar to consulting an expert. Meta Agent has two core capabilities:

    1. Asset inventory: The agent automatically scans and parses metadata to generate key knowledge, such as business descriptions for tables and fields, SQL comments, data lineage, usage instructions, and a business catalog. This knowledge becomes the foundation for powering precise AI services.

    2. Asset Q&A: In Data Copilot, you can use natural language to easily perform various interactive tasks, such as finding assets, analyzing data, and receiving usage recommendations.

    Scenarios

    Meta Agent offers a Database Edition and a Data Lakehouse Edition to address the specific pain points of different roles:

    Target audience

    Core pain points

    Meta Agent solution

    Database administrator

    Spends significant time answering repetitive questions about database schemas, table schemas, and usage standards.

    Automates the inventory and management of database knowledge. This frees the database administrator (DBA) from repetitive Q&A tasks, allowing them to focus on higher-value management work.

    Database/Application developer

    Frequently needs to look up database and table information, write complex SQL queries, and understand business logic.

    You can interact directly with the database in Data Copilot to quickly obtain introductions to databases and tables, generate SQL, analyze errors, and interpret standards. This greatly improves development efficiency.

    Data platform/Data lakehouse administrator

    Faces challenges in uniformly inventorying, describing, and governing large-scale, multi-modal data assets.

    It automates the inventory of all data assets and automatically generates an asset catalog, business descriptions, data lineage, and metric definitions. This significantly reduces data governance costs.

    Data analyst/Data consumer

    Struggles to find, understand, and use data, resulting in low data utilization.

    You can interact securely using natural language to easily find, query, and use data. This significantly lowers the barrier to data consumption.

    Version selection

    To better serve different scenarios, Meta Agent is available in two editions. You can select an edition based on the following table:

    Comparison dimension

    Meta Agent Database Edition

    Meta Agent Data Lakehouse Edition

    Core focus

    Database development and management agent

    Global data asset management and consumption agent

    Scenarios

    Reduces database management costs. Improves business development efficiency and stability.

    Reduces data asset governance costs for the platform. Improves the efficiency of finding, querying, and using data.

    Target users

    DBAs, database developers, and application developers.

    Data lakehouse administrators, data development engineers, data analysts, and business data consumers.

    Core feature differences

    Focuses on generating descriptions and usage instructions for databases, tables, and fields, and on development-oriented Q&A.

    Focuses on in-depth inventory of all assets. It also supports the generation of advanced knowledge such as data lineage, business terms, and metric definitions.

    Detailed feature comparison:

    Feature module

    Features

    Database Edition

    Data Lakehouse Edition

    Data source

    Multi-cloud, multi-modal data sources

    • MySQL: RDS for MySQL, PolarDB for MySQL, other MySQL sources

    • PostgreSQL: RDS for PostgreSQL, PolarDB for PostgreSQL, other PostgreSQL sources

    • SQL Server: RDS for SQL Server, other SQL Server sources

    Supported data lakehouses include the following: AnalyticDB for MySQL, AnalyticDB for PostgreSQL, SelectDB, Starrocks, ClickHouse, MaxCompute, and data warehouse service (DWS)

    Asset Map

    Asset search

    ✅

    ✅

    Business catalog

    ✅

    ✅

    Asset details

    Does not support data lineage, usage instructions, or quality

    ✅

    Asset inventory

    Import documentation

    ✅

    ✅

    Data sampling

    ✅

    ✅

    Code parsing

    ✅

    ✅

    Data lineage generation

    ❌

    ✅

    Business term generation

    ❌

    ✅

    Metric definition generation

    ❌

    ✅

    Usage instruction generation

    ❌

    ✅

    Catalog generation

    ✅

    ✅

    Asset Q&A

    Data Copilot Q&A

    ✅

    ✅

    Note

    To ensure the quality and timeliness of the business knowledge inventory, the Meta Agent service includes a background intelligent analysis task. This task regularly scans, analyzes, and performs inference on your metadata and sample data. The state-of-the-art (SOTA) large language model calls required for this process are a core part of this service. All related token consumption costs are included in the Meta Agent service package that you purchase. You do not need to pay additional fees. For reference, the approximate scale of model calls for the background task is as follows:

    • Database Edition: For each managed instance, a minimum of 8 million SOTA model tokens are used for daily asset inventory and summarization. If the instance contains many data assets, the daily token service limit does not exceed 16 million.

    • Data Lakehouse Edition: For every 1,000 managed tables, a minimum of 200 million SOTA model tokens are used for daily asset inventory and summarization. If the metadata volume is large, the daily token service limit does not exceed 400 million.

    Core advantages

    1. Comprehensive service:
      Meta Agent provides an end-to-end service that ranges from asset inventory and knowledge generation to natural language interaction. It covers the four core stages of data management: managing, finding, querying, and using data.

    2. Precise feedback:
      By deeply understanding the business knowledge generated from the inventory, the agent can provide precise answers that are better aligned with your enterprise's business logic. The knowledge base can also continuously self-optimize based on user feedback.

    3. Open ecosystem:
      The agent's calling capabilities (through APIs or MCP) and the knowledge it generates can be opened for integration and use by other platforms or AI applications. This helps build an extensible intelligent ecosystem.

    4. Secure access:
      All Q&A interactions strictly adhere to the data permission system configured in DMS. This ensures the security and compliance of enterprise data while providing a convenient user experience.

    Limits

    • Currently supported regions: China (Hangzhou), China (Shanghai), China (Shenzhen), China (Chengdu), China (Beijing), China (Zhangjiakou), Singapore, and Malaysia (Kuala Lumpur).

    • Currently supported data sources:

      • MySQL: RDS for MySQL, PolarDB for MySQL, and other MySQL sources.

      • PostgreSQL: RDS for PostgreSQL, PolarDB for PostgreSQL, and other PostgreSQL sources.

      • SQL Server: RDS for SQL Server and other SQL Server sources.

      • Data lakehouses: AnalyticDB for MySQL, AnalyticDB for PostgreSQL, SelectDB, Starrocks, ClickHouse, MaxCompute, and DWS.

    • The instance to be inventoried must be registered in Data Management (DMS). For more information about how to register an instance, see Register an ApsaraDB database and Register a database from another cloud or a self-managed database.

    • When you register a database instance, you must enable security hosting for the instance.

    • The database account used must have query permissions on the target database. For more information about how to view your permissions, see View my permissions.

    Asset inventory

    1. Log on to the DMS console V5.0.

    2. In the top menu bar, choose Data Asset > Asset Map. Alternatively, in the console in simple mode, click the 2023-01-28_15-57-17 icon in the upper-left corner and choose All Features > Data Asset > Asset Map.

      image

    3. (Optional) If you have not purchased Meta Agent, click the Purchase Now button. You can then select a Meta Agent edition and an expansion package as needed.

      image.png

    4. On the page, find the Asset Inventory box and click the Start Inventory button.

    5. On the Instance, Database, or Table tab, select the inventory granularity and then select the checkbox of the target object.

      image

      Note

      We recommend that you perform the inventory at the database or table level. This prevents the process from taking too long if there are many objects to scan.

    6. Click Next to configure the inventory.

    7. Follow the wizard to complete the inventory configuration.

    8. After you confirm the configuration, click the Start Inventory button at the bottom of the page. The system then enters the Knowledge Generation and Confirmation stage.

    9. After the inventory is complete, you can review, edit, and adopt the generated knowledge that is in the Pending Adoption state to make it effective.

      image

      • View and edit knowledge

        • Select the row for the target knowledge and click the Details button.

          Note

          If you only need to view the knowledge details without performing any operations, you can click Cancel after you finish viewing.

        • In the dialog box that appears, click the image icon in the Knowledge Description Comparison or Knowledge Content Comparison section to edit the content.

        • After you finish editing, click the image icon to save the changes.

          Note

          After you save the changes, the adoption status automatically changes to Adopted.

      • Adopt knowledge

        • Adopt individually: Click the Adopt button in the row for the target knowledge.

        • One-click adoption: Click the One-click Adoption button at the top of the list to adopt all knowledge that is in the pending adoption state.

    10. View table details.

      1. Go back to the Asset Map page. In the search box, enter the name of the target table and start the search.

        image

      2. In the search results, click Details to the right of the target table. You can view information about the data table, such as its Basic Information, Detailed Properties, Usage Instructions, Data Lineage, and Knowledge Management. You can manage knowledge on the Knowledge Management tab.

        image

    Asset Q&A

    1. Open DMS Data Copilot.

      Method 1

      1. Go to the Asset Map page. In the Asset Q&A box, click the Asset Q&A button.

        image

      2. In the dialog box that appears, select the target database and then log on to the instance.

      3. After you log on, the DMS Data Copilot dialog box appears.

      Method 2

      1. Go to the DMS home page.

      2. In the navigation pane on the left, double-click the name of the target database in the database instance.

      3. At the top of the SQLConsole tab, click Copilot.

        screenshot_2025-09-23_14-19-43

    2. In the Copilot dialog box, you can ask questions in natural language. For example:

      • "Find tables related to user information."

      • "What fields are in the order table?"

      • "What was the total sales amount last month?"

      Copilot provides precise answers based on your adopted knowledge base. For more information about advanced features, see Data Copilot (New).