Agentic RAG

Introducing the Smart Agent, a new capability in Progress Agentic RAG that breaks complex questions into sub-questions, decides where to look for each one, and keeps searching until the answer is complete. It works on existing setups with no new pipeline or re-indexing. The blog highlights two additions: live data from business apps like Zendesk, Salesforce and ServiceNow through MCP, and a built-in web search tool backed by Perplexity and Google Gemini. Both are cited alongside internal content in a single answer.

The October release adds three capabilities to Progress Agentic RAG with nothing to install: a Smart Agent that answers questions using both indexed content and live business apps like a CRM, a native Microsoft Teams app, and a WordPress plugin that keeps content current. Built-in citations, quality scoring and audit logs let reviewers trace each step the agent takes. The blog ends by inviting readers to an October 22 walkthrough or a demo.

AI models can now take a million tokens in a single prompt, so it’s fair to ask why an agent still needs a layer that hunts down the few relevant passages before it answers. Narrowing first is cheaper and more accurate and it keeps you free to change AI vendors later.

AI token economics connects model usage to the real cost of running an AI application. In this article, we’ll trace token consumption through a RAG pipeline and explore how teams can estimate and manage those costs.