> For the complete documentation index, see [llms.txt](https://documentation.astera.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.astera.com/astera-studio/workspace/connections.md). # Connections {% tabs %} {% tab title="Database" %} ## Database Connection A Database Connection in Astera Studio is a saved link to a database server — SQL Server, PostgreSQL, MySQL, Oracle, BigQuery, Snowflake, and more. Once a database connection is set up in your workspace, AI agents, Datapreps, and Dashboards can read from and write to that database without you needing to re-enter credentials each time. In this article, we will learn how to create a database connection, the providers we support, the fields each provider needs, and how to use the AI-suggested Database Connection cards that Astera Studio surfaces in chat. ### Where Database Connections Live Database connections live in the DB Connections category of the Workspace Panel. Each connection is a saved .sact file in your workspace and is encrypted at rest in the tenant-scoped credential vault — Astera Studio never stores credentials in plain text. ### Supported Database Providers Astera Studio supports more than twenty database providers out of the box. The most common are shown below; the full list appears in the Data Provider dropdown on the connection form. * Microsoft SQL Server (with Windows Authentication support) * PostgreSQL * MySQL and MariaDB * Oracle (ODP.NET and Managed ODP.NET) * Azure SQL Server, Azure Synapse (with Azure AD authentication) * Amazon Aurora MySQL and Aurora PostgreSQL * Snowflake (with S3 or Azure Blob staging) * Google BigQuery * IBM DB2, Netezza, Teradata, Vertica ### Steps to Create a New Database Connection You can create a new database connection in two ways: from the workspace, or from an AI-suggested chat card. The form is the same; only the entry point differs. #### From the workspace 1. Click ***Add Item > Database***.
2. The New Database Connection window will open.
3. Enter a meaningful Connection Name (for example, SalesProdReader). The name must be 80 characters or fewer and cannot contain reserved characters. 4. Pick a Data Provider from the dropdown. The form below adapts to show only the fields the selected provider needs. 5. Fill in the provider-specific fields (server, database, port, schema, credentials). 6. Click **Test Connection** to verify Astera Studio can reach the server with the credentials you supplied. 7. Click Save. The connection appears under DB Connections in the Workspace Panel and is available for AI agents, Datapreps, and Dashboards immediately. #### From an AI-suggested chat card When the AI knows you need to connect to a database — for example, after you mention a production database by name in chat — it can drop a Database Connection card directly into the conversation. The card already knows the provider, host, port, and database name; you only supply the credentials. 1. Describe in chat which database you want to connect with, and the agent creates a suggested card in your chat thread.
2. Review the prefilled summary on the card (provider, host, port, database, authentication type). 3. Click **Connect** (or Reconnect / Set up, depending on the card kind). The Database Connection editor opens with every field already populated.
4. Enter the credentials Astera Studio asks for — typically a username and password, or an Azure AD token. 5. Click **Test Connection**, then **Save**. ### Provider-Specific Fields Each provider shows the fields it needs and hides the ones it doesn't. The two most common shapes are shown below. #### SQL Server * Server Name — host name or IP address (required). * Database — pick from the dropdown, or type the name. Click Load Databases to populate the dropdown using the credentials you have entered. * Schema — optional, defaults to dbo. * Port — defaults to 1433. * Use Windows Authentication — checkbox. When ticked, Astera Studio uses the currently logged-in Windows user and hides the User / Password fields. * User ID and Password — required if Windows Authentication is not used. #### PostgreSQL * Server Name — host name or IP address (required). * Database — name of the database (required). * Schema — optional, defaults to public. * Port — defaults to 5432. * User ID and Password — required. #### Azure SQL Server Azure SQL Server adds an Azure-specific Authentication dropdown above the credentials fields. Pick one of: * SQL Server Authentication — classic username / password. * Azure Active Directory — MFA — interactive sign-in with multifactor authentication. * Azure Active Directory — Password — username and password against Azure AD. * Azure Active Directory — Integrated — uses the currently logged-in Windows account. For the AD modes, Astera Studio collects Client ID, Redirect URI, and (depending on the mode) exchanges a token for you on submit. ### How Credentials Are Stored Credentials you enter in the connection form never leave Astera Studio in plain text. On save, they are encrypted with a tenant-scoped key and stored in the Credential Vault (DB-backed). The Workspace Panel and chat surfaces only ever see the connection name — never the password. If you need to update a credential (for example, after a password rotation), open the connection from the Workspace Panel, type the new value, and save. The vault entry is updated in place. {% endtab %} {% tab title="API" %} ## API Connection An **API Connection** in Astera Studio is a saved configuration for a REST API or web service — its base URL, headers, parameters, and authentication. Once an API connection is set up in your workspace, AI agents, Datapreps, and Dashboards can call the API by name, and credentials are handled by the platform instead of being passed around in chat. In this article, we will learn how to create an API connection, the authentication schemes we support, how to use the AI-suggested API Connection cards in chat, and how Astera Studio's Test Connection feature gives you confidence before you save. ### Where API Connections Live API connections live in the API Connections category of the Workspace Panel. Each connection is a saved .sact file in the workspace; the secrets it contains (API keys, client secrets, bearer tokens) are encrypted and rest in the tenant-scoped Credential Vault. #### Steps to Create a New API Connection 1. Click ***Add Item > API Connection.***
2. The API configuration window will open. 2. Enter a Connection Name (for example, StripeBilling or SalesforceProdAPI). 3. **Basics:** * Base URL - Here, you can specify the base URL of the API which will prepend as a common path to all API endpoints sharing this connection. A Base URL usually consists of the scheme, hostname, and port of the API web address. * Timeout - Specify the duration, in milliseconds, to wait for the API server to respond before giving a timeout error. 4. **Authentication:** * Security Type - Specify the authentication type for the API. Astera supports the following authentication types.
5. **Default Parameters:** * Add Parameter - This is where you can define query or header parameters to be shared across all clients using the same connection.
6. **Advanced:** * Include Client SSL Certificate *-* Check this box to include an imported client certificate for the specified base URL. * Enable Authentication Logs *-* Select this checkbox to enable authentication logging for APIs.
7. Click **Test Connection**. Astera Studio sends a sample request using the values you have entered and shows the result. 8. Click **Save**. ### Configuring API from Chat The AI can suggest an API Connection when it determines you need one — for example, after you describe an external service you want to integrate. The API Connection card in chat already knows the base URL, the auth shape, and the typical scopes; you supply the secret. 1. Describe in chat what API you want to connect with, and the agent creates a suggested card in your chat thread.
2. Review the prefilled summary (base URL, authentication type, headers, OAuth flow, scopes). 3. Click **Connect.** The API Connection editor opens with every field already populated.
4. Enter the credentials Astera Studio asks for — usually just an API key, bearer token, or OAuth client secret. 5. Click **Test Connection**, then **Save**. {% endtab %} {% tab title="Connector" %} ## Connectors A **Connector** in Astera Studio is the bridge that lets AI agents act on the world outside the platform — Google Workspace, Slack, GitHub, your internal services, anywhere. Connectors are built on the Model Context Protocol (MCP), an open standard for AI tool integrations. Astera Studio offers connectors in two ways: a curated **Connector Marketplace** with pre-built integrations you can install in a few clicks, and **Custom Connectors** that let you point Astera Studio at any MCP-compatible server you run yourself. In this article, we will learn how to browse and install marketplace connectors, how to set up custom connectors against your own MCP servers, how the AI suggests connectors in chat, and how connector health is continuously monitored. ### Where Connectors Live Connectors appear in two places: * In the Connector Marketplace — a dedicated catalog you open from the Workspace Panel. This is where you browse, install, claim, and configure connectors. * In the Connectors category of the Workspace Panel — once a connector is installed, it appears here next to your DB and API connections. ### Browsing the Connector Marketplace The marketplace organizes connectors into categories so you can find what you need quickly. * **All connectors** — every connector available to you. * **Your connections** — connectors you have already installed and configured. * **Shared by your org** — connectors set up by someone else in your tenant. You can claim these to use them with your own credentials. * **Browse all** — the full catalog of installable connectors. * **Categories** — Productivity, Development, Communication, Data, Automation, and more. Each category shows the number of connectors in it. Use the search bar at the top of the catalog to find a connector by name. Use the filter bar to narrow by connection status — for example, *Connected, Not connected, Reconnect needed*. ### Steps to Install a Marketplace Connector Most marketplace connectors use OAuth and take less than a minute to install. 1. Click a connector tile in the marketplace to open its detail page. 2. Click the primary action button — usually labelled Connect, Set up, or Claim, depending on the connector's state. 3. If the connector uses OAuth, a popup opens with the provider's consent screen. Sign in and approve the requested scopes. 4. If the connector uses a bearer token, paste the token into the field provided. 5. Astera Studio stores the credential in the tenant-scoped Credential Vault and runs an initial health probe to confirm the connection works. 6. When the probe succeeds, the connector's status changes to Connected and a green checkmark appears on its tile. ### Custom Connectors If the integration you need is not in the marketplace, you can connect to any MCP-compatible server you run yourself. This is what we call a Custom Connector. A Custom Connector tells Astera Studio: * Where your MCP server lives (the URL). * How to talk to it (the transport — Server-Sent Events, Streamable HTTP, or Standard I/O). * How to authenticate (no auth, bearer token, basic auth, API key, or OAuth 2.0). #### Steps to Add a Custom Connector 1. In the Connector Marketplace, click Connect new MCP server at the bottom of the left rail. 2. Enter the MCP Server URL — for example, ; 3. Enter a Display Name so you can identify the connector in your workspace. 4. Choose the Transport Type: Server-Sent Events (SSE) (most common), Streamable HTTP, or Standard I/O for local stdio-based MCP servers. 5. Pick the Authentication type from the dropdown — None, Bearer, Basic, API Key, or OAuth 2.0. The form below adapts to show the fields each scheme needs. 6. If you chose OAuth 2.0, decide whether to use Dynamic Client Registration (DCR). With DCR enabled, Astera Studio auto-discovers the OAuth endpoints from your server's .well-known/oauth-authorization-server document. Without DCR, fill in the Authorize URL, Token URL, Client ID, Client Secret, and Scopes manually. 7. Optionally, enable PKCE (Proof Key for Code Exchange) — recommended for OAuth flows. 8. Click Connect. Astera Studio establishes the MCP session, discovers the tools your server exposes, and lists them in the connector's detail page. From a connector's detail page, you can manually run a fresh probe (the Discover tools button), view the list of tools your server exposes, and see the refresh-token status if the connector uses OAuth. ### Connection Scope: Private vs Tenant-Shared By default, every connector you install is private — only you can use it, and only your credentials are stored. If you want the rest of your tenant to use the same connector, you can share it. 1. Open the connector's detail page from the marketplace. 2. Click the ⋮ (more) menu in the header and choose Share. 3. In the Share Connection dialog, toggle on Share with \[your org]. 4. Choose whether the connector should be shared as a Full Connection (everyone uses your credentials) or as a Custom Configuration (everyone has to supply their own credentials). 5. Click Save. The connector becomes visible in the Shared by your org section for other users in your tenant. **Important:** Full Connection sharing is convenient but means every user inherits the same access. Use it for read-only integrations or for service accounts. Use Custom Configuration when each user should authenticate as themselves. ### AI-Suggested Connector Cards in Chat When the AI realises you need a tool that is not yet connected, it surfaces a Connector card directly in chat — the same way it surfaces Database and API offers. The card already knows the connector name, the suggested transport, and the auth shape; you fill in the credentials. The cards come in four flavours: * Reconnect — an existing connection has expired; one click takes you to the Re-authenticate step. * Set up (Claim) — a tenant-shared connector exists; one click opens the claim flow so you can use it with your own credentials. * Connect (Marketplace) — a marketplace connector matches your need; one click opens its detail page in the marketplace. * Connect new MCP server (Custom) — no marketplace match; one click opens the Custom Connector flow with the URL, transport, auth type, and scopes already filled in. ### Discovering and Using Tools Every connector exposes a set of tools — discrete actions the AI can call. After you install a connector, click its tile or item to open the detail page; the tool list appears near the top. Tools are automatically available to chat agents in the same workspace. You can also bind a Skill to a specific tool list — see the AI Skills documentation for that workflow. {% endtab %} {% endtabs %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://documentation.astera.com/astera-studio/workspace/connections.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.