Build Agents in Relevance AI with Access to Live PostgreSQL Data via CData Connect AI

Relevance AI is an AI automation and agent-building platform that enables organizations to create autonomous workflows powered by natural language reasoning. Users can visually design agents that interact with APIs, databases, and third-party systems to complete everyday business tasks or data operations.

By integrating Relevance AI with CData Connect AI through the built-in MCP (Model Context Protocol) Server, your agents can query, summarize, and act on live PostgreSQL data in real time. This connection bridges Relevance AI intelligent workflow engine with the governed enterprise connectivity of CData Connect AI ensuring every query runs securely against authorized sources without manual data export.

This article outlines the steps to configure PostgreSQL connectivity in Connect AI, register Connect AI in Relevance AI, and build an agent that interacts with live PostgreSQL data.

Step 1: Configure PostgreSQL Connectivity for Relevance AI

Connectivity to PostgreSQL from Relevance AI is made possible through CData Connect AI's Remote MCP Server. To interact with PostgreSQL data from Relevance AI, we start by creating and configuring a PostgreSQL connection in CData Connect AI.

1. Log into Connect AI, click Sources, and then click Add Connection
👁 Adding a connection in Connect AI

2. Select PostgreSQL from the Add Connection panel
👁 Selecting data source

3. Enter the necessary authentication properties to connect to PostgreSQL.

   To connect to PostgreSQL, set the Server, Port (the default port is 5432), and Database connection properties and set the User and Password you wish to use to authenticate to the server. If the Database property is not specified, the data provider connects to the user's default database.

   SSH Connectivity for PostgreSQL

   You can use SSH (Secure Shell) to authenticate with PostgreSQL, whether the instance is hosted on-premises or in supported cloud environments. SSH authentication ensures that access is encrypted (as compared to direct network connections).

   SSH Connections to PostgreSQL in Password Auth Mode

   To connect to PostgreSQL via SSH in Password Auth mode, set the following connection properties:

   • User: PostgreSQL User name
   • Password: PostgreSQL Password
   • Database: PostgreSQL database name
   • Server: PostgreSQL Server name
   • Port: PostgreSQL port number like 3306
   • UserSSH: "true"
   • SSHAuthMode: "Password"
   • SSHPort: SSH Port number
   • SSHServer: SSH Server name
   • SSHUser: SSH User name
   • SSHPassword: SSH Password

   SSH Connections to PostgreSQL in Public Key Auth Mode

   To connect to PostgreSQL via SSH in Password Auth mode, set the following connection properties:

   • User: PostgreSQL User name
   • Password: PostgreSQL Password
   • Database: PostgreSQL database name
   • Server: PostgreSQL Server name
   • Port: PostgreSQL port number like 3306
   • UserSSH: "true"
   • SSHAuthMode: "Public_Key"
   • SSHPort: SSH Port number
   • SSHServer: SSH Server name
   • SSHUser: SSH User name
   • SSHClientCret: the path for the public key certificate file
   👁 Configuring a connection (Salesforce is shown)
   
4. Click Save & Test
5. Navigate to the Permissions tab and update user-based permissions
👁 Updating permissions

Add a Personal Access Token

A Personal Access Token (PAT) is used to authenticate the connection to Connect AI from Relevance AI. It is best practice to create a separate PAT for each integration to maintain granular access control.

1. Click the gear icon () at the top right of the Connect AI app to open Settings
2. On the Settings page, go to the Access Tokens section and click Create PAT
3. Give the PAT a descriptive name and click Create
👁 Creating a new PAT

4. Copy the token when displayed and store it securely. It will not be shown again

With the PostgreSQL connection configured and a PAT generated, Relevance AI can now connect to PostgreSQL data through Connect AI.

Step 2: Configure Connectivity in Relevance AI

The CData Connect AI MCP endpoint and authorization details are registered within Relevance AI so that agents can call live data from Connect AI.

1. Sign in to Relevance AI and create an account if you do not already have one
2. From the sidebar, navigate to Agents and then click on New Agent
👁 Creating a new agent in Relevance AI

3. Select Build from scratch and name the agent (eg; CData MCP Server)
👁 Building an agent from scratch

4. Inside the agent editor, select Advanced and then switch to the MCP Server tab
👁 Opening MCP Server settings

5. Click + Add Remote MCP Tools
6. In the dialog that appears, fill out the fields as follows:
• URL: https://mcp.cloud.cdata.com/mcp
• Label: Any custom label (eg; cdata_mcp_server)
• Authentication: Select Custom headers
• Add header key:value pair. Combine your email and PAT as email:PAT and encode that string in Base64 and then prefix with the word Basic
  • Key: Authorization
  • Value: Basic base64(email:PAT)
👁 Connecting to CData Connect AI MCP Server in Relevance AI

Click Connect to establish the connection. Relevance AI will verify your credentials and register the CData Connect AI MCP Server for use in agents.

Step 3: Build and Run a Relevance AI Agent with Live PostgreSQL Data

1. Switch to the Run tab for your agent
2. Enter a task for example, "List the five most recent incidents from ServiceNow"
👁 Running the Relevance AI agent

3. The agent will query Connect AI via the MCP endpoint and display live results from PostgreSQL data
👁 Example query result from Connect AI

With the connection complete, Relevance AI agents can now issue queries, retrieve records, and perform AI-driven tasks over live PostgreSQL data through CData Connect AI MCP Server.

Get CData Connect AI

To access hundreds of SaaS, Big Data, and NoSQL sources directly from your cloud applications, try CData Connect AI today!