VS Code extension for converting Natural Language Queries to SQL using OpenAI's API.
Users can supply a database schema and a question to convert, and the extension will generate SQL that can be run on the database. Users use their own OpenAI API key and select a model.
[!WARNING]
This extension is in beta and is subject to bugs and incomplete features. Please open an issue if you find a bug or have a feature request.
Go the the extension's settings, enter your API key, and select a model.
Use the NLQ-to-SQL: Add Database command to add a database to the extension
Use the NLQ-to-SQL: Generate SQL w/ AI command to generate SQL from a question
Features
Views
This extension contributes a new view container called NLQ-to-SQL to the Activity Bar with the following views:
Database Explorer: Shows the databases you've added to the extension. You can add a database by clicking the + button in the view's title bar. In this view you can manage and enable table contexts to use for NLQ-to-SQL queries.
Active Table Context: Shows the tables included in the active table context. Any NLQ-to-SQL queries will use these tables only to generate SQL, instead of sending the entire database schema to OpenAI's API.
Generated SQL: Shows a history of all NLQ-to-SQL queries you've generated. From here you can copy the SQL to your clipboard or insert it into the editor.
NLQ-to-SQL: Generate SQL w/ AI: Input a question and generate SQL using OpenAI's API. The resulting SQL will be inserted into the editor at the current cursor position.
NLQ-to-SQL: Add Database: Follow the instructions to add a database to the extension. You must have a database added to use NLQ-to-SQL.
NLQ-to-SQL: Clear Query History: Clear the history of generated SQL queries.
Note:
You must have an OpenAI API key to use this extension. You can get one here.
This extension does not execute the generated SQL. You must run the SQL on your database yourself. There are a number of VS Code extensions that can do that for you, such as SQLTools.
All database information and other extension information is associated with a specific Workspace.
