Pega Agents become more powerful when they can connect with external tools, services, and information beyond the application context.
In this demo, I explored how Model Context Protocol (MCP) can be used to extend Pega Agent capabilities through MCP server integration. With this approach, the Agent can access and use tools exposed by the MCP server whenever additional information or external action is needed.
Demo video: https://youtu.be/b0ys3F1hCHc?si=tgel-7vAkglBJJIb
Medium article: https://medium.com/@stellarnexus/extending-pega-agent-capabilities-using-mcp-server-integration-99553b4790c2?sharedUserId=stellarnexus
This helps create Agent experiences that are not limited only to the data and logic available inside the application, but can also interact with connected systems in a more flexible way.
The execution flow can also be validated using AI Tracer, which makes it easier to understand how the Agent identifies the need for a tool, invokes it, and continues the overall flow.
This is a simple but useful step toward building more dynamic, connected, and intelligent Agent experiences in Pega.