Agent

Overview

Agents communicate with an LLM with a given clear role and purpose, and equipped with specific tools to interact with the platform. Agents also understand context and have access to your RAG knowledge bases.

Agents are added to a team where they are executed in a given order and handle inputs and outputs passed between agents in a team.

Example agent

For an example of how to set up an agent refer to How to create a simple agent.

Choice of LLM

Platform version 5.2 introduced the ability to define an agent using your choice of LLM provider. The current two options are:

• OpenAI
• Anthropic

For more information, refer to Support for different LLMs.

System and custom agents

There are two types of AI Agent:

• System Agent: Pre-built on the Platform for easy invocation
• Custom Agent: You can create your own custom agent

Click on either agent type to see more information.