What are background agents?

Learn about background agents in AI Chatbot Hub: automated assistants that work behind the scenes to enhance user experience.

AI Chatbot Hub provides two types of Agents:

• User-facing: This Agent interacts directly with users in a conversational Q&A manner. Only one user-facing Agent is active at a time when a new query is entered into the chatbot.

• Background: These Agents do not interact directly with users but monitor conversations in real-time. All background Agents activate when a user submits a new query to the chatbot.

Unlike user-facing Agents, background Agents are designed to observe conversations and provide support by setting tags or calling functions when specific conditions are met. Each Agent functions independently, so you cannot influence one Agent’s behavior, selection priority, or bias from another.

Background Agents continuously run whenever a user enters a new query. This setup allows you to establish “multi-threaded” workflows, enhancing your chatbot’s accuracy and consistency. Rather than having a single Agent manage multiple independent tasks, specialized background Agents can observe and act on specific conversation events in real-time, producing better outcomes.

When configuring a background Agent, keep the prompt simple and focused. Since these Agents don’t generate conversational output, instructions should be along the lines of “monitor the conversation for specific behaviors or intents.” You can direct the background Agent to assign tags or trigger functions as certain conditions are met.

In some scenarios, you might want to enable RAG (Retrieval-Augmented Generation) for background Agents—for instance, if they need to check a user’s statement against specific policies or historical data before tagging. To grant such access, go to the “Knowledge” tab in the Agent’s edit settings.

A typical use for background Agents is tagging conversations.