Create and Manage a Connection

A connection is a saved configuration that lets PlaidCloud reach an external system — a database, data lake, ERP, REST API, cloud service, or Git provider. Workflow steps reference the connection by name, so endpoint details and credentials live in one place and are reused across every step and project that needs them.

Connections are managed from Tools > Connections.

Create a Connection

1. Open Tools > Connections
2. Click New Connection in the toolbar and choose the system you want to connect to from the menu
3. Fill in the connection form (the tabs and fields vary by type — see Connection Form below)
4. Click Create to save

The new connection appears in the list, owned by you.

Note

The list of connection types and the fields each one needs are cataloged in the Connectors reference. For a service without a dedicated connector, use the Generic REST Connection.

Connection Form

Most connections open a multi-tab form. The exact tabs depend on the type, but you’ll commonly see:

• Connection Config (or Connection Details for REST) — the connection Name, an optional Memo, the host/port/database (or host/auth type for REST), the Usage toggle (Active), and the Security Model. On a database connection the credentials sit here in an Auth Credentials group; on a REST connection they get their own Authentication tab.
• SSL Config — Use SSL plus the SSL mode and optional client/root certificates for an encrypted connection.
• SSH Config — Use SSH plus the bastion host, port, user, and key to tunnel the connection through an SSH jump host.

Password and token fields are write-only: they show blank when you reopen a saved connection, and leaving them blank on save keeps the stored secret. Retype a value only when you want to change it.

Required fields are validated on save; if something is missing, the form switches to the tab holding the first invalid field and flags it.

Sandbox and Production

Every connection editor carries a Sandbox (non-production) connection checkbox with a live badge next to it. Tick the box when the connection points at a non-production instance — a sandbox ERP, a staging database — and the badge reads SANDBOX (green); leave it clear and the badge reads PRODUCTION (red). The flag is saved with the connection and labels it only; it doesn’t change where the connection points. Use it so anyone opening a connection can tell at a glance whether it targets a live system before they run a step against it.

Security Model

The Security Model on a connection controls who in the workspace can use it:

Option	Who can use the connection
Private (Only Owners)	Just the owners.
Specific Members Only	Owners plus an explicit list of members.
Specific Security Groups Only	Owners plus members of named security groups.
All Workspace Members	Everyone in the workspace.

When you choose Specific Members Only or Specific Security Groups Only, grant the access from the list’s Actions menu — Members, Groups, Agents, and Owners each open a dual-list picker for adding and removing access.

Environments

A connection can hold a different configuration per environment (for example a development host and a production host under one connection name). The environment selector in the toolbar chooses which environment’s configuration you are viewing and editing.

• New Environment — add a connection environment
• Edit Environment — rename the selected environment
• Delete Environment — remove the selected environment

The Available column in the list shows whether the selected connection is configured for the chosen environment.

Note

Saving a configuration merges into the stored one for that environment, so blank write-only secrets are never overwritten with empty values.

Environments are shared with PlaidLink agents, so a project’s environment can switch the agent that runs its remote steps as well as the connections it reads. See Agent Environments.

Edit a Connection

1. Select the connection in the list
2. Click Edit in the toolbar (or right-click the row)
3. Change the fields and click Update

Test a Connection

How you test depends on the type:

• Database connections — Databricks, Snowflake, PostgreSQL, and Redshift — have a Test Connection button in the editor. It opens the connection with the credentials you entered and reports success (with the server version it reached) or the reason it failed, without closing the editor, so you can fix and retry in place.
• Database connections with SSH also expose a test for the SSH tunnel in the SSH Config tab.
• The Generic REST Connection has a Testing tab that fires an ad-hoc request through the saved connection.

Save the connection before testing — tests run against the stored configuration.

Note

Test Connection is for the database connection types above. NetSuite and other REST connections don’t have it — smoke-test those by sending a small request from a workflow step (for NetSuite, see Connect to NetSuite).

Clone, Delete, and Trace Usage

• Clone — duplicate a connection’s configuration as the basis for a new one. See Clone a Connection.
• Delete — remove the selected connection from the toolbar (you are asked to confirm).
• Where Used… — list the workflow steps that reference the connection, so you can see the impact before editing or deleting it.

Caution

Edit, Clone, and Delete are only available on connections you own. If those actions are greyed out for the selected row, you are not an owner — ask an owner to make the change or to add you via Owners.

Related

• Connectors reference — the full catalog of supported systems and their fields.
• Clone a Connection
• Singer Sources