# Azure SQL Developer Hub

> The front door for building applications on Azure SQL Database with an AI coding agent. Start free in the cloud, hand your agent a build prompt, and verify a working app. Built for AI workloads: native vector search, in-engine embeddings, JSON, and row-level security. Agent skills teach Claude Code, GitHub Copilot, Codex, and Cursor how the engine works.

_Last updated: September 2026._

> Every page on this site has a markdown twin at the same path with `.md`. The whole site in one file is at /llms-full.txt.

## Documentation
- [Get running](https://microsoft.github.io/azure-sql-dev-hub/#get-running): create a free Azure SQL Database, verify the connection, pick a path.
- [Skills](https://aka.ms/azuresql-skills): what the agent skills do, per-tool install, and the full catalog.
- [Before you build](https://microsoft.github.io/azure-sql-dev-hub/#faq): what you need, what it costs, whether you need the container, whether it is safe to let an agent write to your database.
- [For agents](https://microsoft.github.io/azure-sql-dev-hub/for-agents.md): how this site is laid out for machine readers.

## Build prompts
Copy-and-run prompts to hand your AI coding agent. Each targets Azure SQL Database in the cloud and ends in a working app, with validation rules at the end:
- [Build a task-tracking app](https://microsoft.github.io/azure-sql-dev-hub/build/javascript-app.md): JavaScript. Build a web app that displays tasks stored in Azure SQL. Start with sample data and make it your own.
- [Build an API for your app](https://microsoft.github.io/azure-sql-dev-hub/build/python-api.md): Python. Give your app a way to create, complete, and retrieve tasks. Build a Python API that connects to Azure SQL with Microsoft Entra, no passwords.
- [Search your content by meaning](https://microsoft.github.io/azure-sql-dev-hub/build/rag-app.md): Python / AI. Find relevant text with vector search in Azure SQL. Build the retrieval layer for a RAG app using sample text and an embedding model you provide.
- [Create a serverless task API](https://microsoft.github.io/azure-sql-dev-hub/build/serverless-api.md): .NET. Read and save tasks through an Azure Functions API backed by Azure SQL. Build and test the functions locally.
- [Respond to data changes](https://microsoft.github.io/azure-sql-dev-hub/build/event-driven-app.md): .NET. Extend the serverless API with a function that reacts when tasks change in Azure SQL. Log inserts, updates, and deletes as a starting point for automation.
- [Separate each customer's data](https://microsoft.github.io/azure-sql-dev-hub/build/multi-tenant.md): JavaScript. Extend a task app for multiple customers. Add row-level security and test that one tenant cannot read another tenant's tasks.

## Key facts
- Target: Azure SQL Database in the Microsoft Azure cloud. The free offer covers every prompt on this site, within its limits. Connect with Microsoft Entra; never a password in code or config.
- Authentication by driver: `mssql` (Node) uses `authentication: { type: "azure-active-directory-default" }`; `Microsoft.Data.SqlClient` (.NET) uses `Authentication=Active Directory Default`; `mssql-python` acquires a token with `azure-identity` and passes it to the driver.
- AI-native: native `VECTOR` type, `VECTOR_DISTANCE` similarity search, `AI_GENERATE_EMBEDDINGS`, `CREATE EXTERNAL MODEL`, native JSON, RegEx. Insert vectors as JSON arrays cast through `NVARCHAR(MAX)` first, or a long embedding is sent as ntext and fails with error 529.
- Event-driven: the Azure Functions SQL trigger requires Change Tracking on the database and the table. Output bindings upsert with `MERGE` and need a primary key and compatibility level 130 or higher.
- Multi-tenant: row-level security with a filter and block predicate keyed on `SESSION_CONTEXT`, set per connection with `sp_set_session_context`.
- The free offer runs on serverless compute with auto-pause; the first connection to a paused database can return error 40613 and succeeds on retry.
- Local development: the Azure SQL Database container runs the same engine offline. Preview; sign up at https://aka.ms/sqldbcontainerpreview-signup. Its prompts are local-scoped and live at https://microsoft.github.io/azure-sql-database-container/.

## Optional
- [Repository](https://github.com/microsoft/azure-sql-dev-hub), short link https://aka.ms/azuresql-hub
- [Install the agent skills](https://aka.ms/azuresql-skills): `npx skills add microsoft/microsoft-sql` (Claude Code, Codex, Cursor, VS Code with GitHub Copilot); Claude Code plugin with `claude plugin marketplace add microsoft/microsoft-sql` then `claude plugin install microsoft-sql@microsoft-sql`; or `gh skill install microsoft/microsoft-sql --all --agent <claude-code|github-copilot|cursor>`.
- [Azure SQL Database container](https://microsoft.github.io/azure-sql-database-container/): the local engine, Preview.