Skip to content

Agents – Introduction

Agents are the central element of the Evoya platform. They form the foundation for intelligent chats, automation, and complex AI-supported workflows.

They combine an AI model with clear instructions, additional knowledge, and optional capabilities to specifically handle tasks and support users.

What is an Agent?

An agent is a configurable AI assistant based on a Large Language Model (LLM) and set up for a specific purpose.

An agent can, for example:

  • Answer questions
  • Summarize information
  • Analyze content
  • Guide users through processes
  • Collaborate with other agents

How an agent behaves depends entirely on its configuration.

What are Agents Used For?

Agents are deployed wherever recurring tasks, information queries, or structured interactions need to be automated or supported.

Typical areas of use include:

  • Website chats
  • Internal assistants for teams
  • Knowledge access to documents and content
  • Analysis and research tasks
  • Multi-step AI workflows

Structure of an Agent

An agent consists of several components. These interact with each other and together determine the agent's behavior:

Intelligence (AI Model)
The AI model is the agent's "brain". It determines how the agent understands, processes, and generates content.
→ More on Intelligence (AI Models)

Instructions (Agent Instructions / Prompt)
The agent instructions describe the agent's role, tasks, style, and limits.
→ More on Instructions (Prompt)

Knowledge (Data Sources)
Through data sources, an agent can access additional knowledge, for example from documents or websites.
→ More on Knowledge (Data Sources)

Capabilities (Tools)
Tools extend an agent's possibilities, such as through searching, reading websites, or analyzing images.
→ More on Capabilities (Tools)

Collaboration (Agent-to-Agent)
Agents can collaborate with other agents and divide tasks among themselves.
→ More on Collaboration (Agent-to-Agent)