MCP in AI Agent Action is live

MCP support is now available in AI Agent Action

AI Agent Action workflows can now connect to MCP sources, giving your agent access to external tools and services during workflow execution. You add the MCP connection in the workflow interface, then use the connected tools to expand what the agent can do inside a flow.

What changed

You can now add MCP connections directly in the AI Agent Action workflow interface. Once connected, the agent can work with MCP-enabled tools instead of being limited to built-in actions alone.

This gives you a practical way to extend agent workflows with external systems that already expose useful capabilities through MCP.

Why it matters

MCP support makes AI Agent Action workflows more flexible when your process depends on information or actions outside ACEIRT™ Fusion. You can connect the workflow to the services your team already uses and let the agent retrieve data or trigger actions in context.

That means fewer manual handoffs and fewer gaps between the workflow and the systems it needs to reach.

How it works

1. Open your AI Agent Action workflow.

2. Add an MCP connection in the workflow interface.

3. Connect the external tool or service you want the agent to use.

4. Run the workflow and use the MCP-enabled actions that become available through the connection.

Once the connection is active, the agent can use capabilities exposed by tools such as search engines, browsers, databases, and APIs.

What to expect

MCP connections are designed to fit into the workflow experience you already use. After you add a connection, the agent can reach supported external capabilities without changing the overall shape of the workflow.

If your workflow needs outside data or action access, MCP gives you a direct path to add it.