For the complete documentation index, see llms.txt.
Skip to main content
Version: 8.7

AI agents

Build and integrate AI agents into your end-to-end processes.

About AI agents

An AI agent is an addressable execution of an LLM-driven loop with shared memory context across iterations. An agent runs a loop where the model decides what to do next, which tools to invoke, and when to stop. The loop is what makes it an agent. A standalone LLM call with no loop and no autonomous tool selection, such as a single connector call that returns output along a fixed execution path, is not an agent.

AI agents can perform a variety of functions, including making decisions, solving problems, interacting with external environments, and taking actions.

AI agent integration features

Use the following Camunda 8 features to integrate AI agents into your processes:

TypeFeatureDescription
BPMN subprocessAd-hoc sub-processA special kind of embedded subprocess with an ad-hoc marker that allows a small part of your process decision-making to be handed over to a human or agent.

Get started with building Your First AI Agent in Camunda.

tip

Explore the version 8.8 documentation to learn more about the additional AI agent features available in Camunda 8.8, such as the new AI Agent connector.