Agents overview

Agents are best described as Natural Language Understanding (NLU) modules. These modules can be included in your app, website, product, or service and translate text or spoken user requests into actionable data. This translation occurs when a user's utterance matches an intent within your agent.

Figure 1. Example of how Dialogflow's fulfillment calls an external API to construct the response to a user.

The matched intent then delivers a response back to the user. This response can be a simple text or spoken acknowledgment or a webhook response that includes information obtained from another system. In Figure 1, Dialogflow sends a webhook request with the location and date parameters to a third-party weather service. This weather service returns a webhook response in JSON format. The agent's custom fulfillment parses the JSON data and delivers a response to the user with the relevant information.

What's next?

Now that you know what an agent is, check out the following pages for more information:

  • Create and manage: See the process of creating a new agent and the various related settings.
  • Machine learning: Learn about the machine learning that's behind an agent and how this helps your agent understand users' requests.
  • Share your agent: See how to collaborate with others on your agent.
  • Make your agent multilingual: Learn how to add languages to your agent, making it accessible world wide.