1. Open the data source list
In the left main navigation, click More, then click Data Sources in the data workspace sidebar to open the data source list. The list has a search box to filter by name and an Add Data button in the top right. Click Add Data to start the creation flow.2. Select a data type
The first step of the creation flow is selecting a data type. AskTable supports 30+ data sources across three categories — File, Database, and Big Data. The full list: Files Excel (.xlsx / .xls), CSV, Feishu Bitable, Workbook (a built-in writable dataset, Beta)
Relational databases
MySQL, PostgreSQL, Oracle, Microsoft SQL Server, TiDB, OceanBase, PolarDB (MySQL / PostgreSQL), Greenplum
Chinese domestic databases
Dameng (DM), KingbaseES, GBase (8a / 8c), XuGu, YashanDB, MogDB, Huawei GaussDB, Tencent TDSQL (MySQL / PostgreSQL)
Data warehouses & big data
ClickHouse, StarRocks, Apache Doris, SelectDB, Apache Hive, Alibaba Cloud MaxCompute, Hologres, Databend, Alibaba Cloud AnalyticDB (MySQL / PostgreSQL), GaussDB(DWS)
Version compatibility: AskTable connects to each engine through SQL dialect adaptation and is broadly compatible across common versions — no specific version required. Tested versions include MySQL 8, PostgreSQL 16, Oracle 11g, SQL Server 2022, plus StarRocks, Apache Doris, SelectDB, Databend, and AnalyticDB for MySQL. Dameng requires the Enterprise edition.
3. Configure the connection
After choosing a database type, fill in the connection details:| Field | Description |
|---|---|
| Data name | Optional. Leave blank to auto-generate. |
| Host | Database server address |
| Port | Pre-filled with the engine’s default; adjust as needed |
| Username / Password | Database credentials |
| Database | The database name to connect to. Some engines (such as MySQL) accept multiple databases separated by commas; for Oracle, enter the Service Name here. |
Fields vary slightly by engine — for example, Oracle uses a Service Name, and Databend requires a warehouse. The form adjusts automatically based on the selected engine.
4. Add data tables
Once connected, AskTable lists the tables and views in the database. Select the tables you want to import and click Confirm Selection, or click Skip to choose them later from the data overview page. This step also offers a Sync Value Index option, which builds an index over field values to improve value-level retrieval accuracy. Building takes time, so enable it only when the dataset is small or retrieval precision matters; you can also trigger it manually after creation.File upload
If you choose an Excel / CSV file type, the second step becomes a file upload page:- Enter a data name (optional)
- Drag and drop or click to upload
.xlsx/.xls/.csvfiles. Multiple files are supported (up to 10; Excel files up to 10 MB each, CSV up to 20 MB). - Click Add Now to finish
Next steps
To learn more about managing and using this data, refer to:- Managing Data — view and manage connected data sources
- Table and Field Notes — add business descriptions to improve AI accuracy