What Are Custom Agents in Moterra?

This article explains what custom agents are and how they help you create tailored AI experiences in Moterra. Custom agents allow admins to configure specialized versions of the assistant by defining their behavior, available tools, and access permissions.

Written By Kristė Vagnerytė

Last updated 11 days ago

What Are Custom Agents Used For?

Custom agents are used to handle specific tasks or serve specific teams.

Instead of relying on one general assistant, you can create custom agents that:

  • Follow predefined instructions

  • Use only selected tools

  • Are available only to certain users or groups

This helps ensure more consistent and predictable results.


What You Can Configure in a Custom Agent

When creating a custom agent, you can configure the following:

Identity

  • Agent name and description

  • Icon

  • AI model used by the agent

  • Which users or groups can access the agent

Instructions

  • The system prompt that defines how the agent behaves

  • What the agent should focus on

  • Any rules or constraints it should follow

Tools

  • Which tools the agent is allowed to use

  • Examples include:

    • Knowledge base access

    • Web search

    • Document creation

    • Image generation

    • Data analysis

All custom agents currently use the same knowledge base, but their behaviour depends on the instructions, tools, and access settings you define.


How to Create a Custom Agent

To create a new custom agent:

  1. Open the Admin portal area in Moterra

  2. Go to Custom agents

  3. Click Create Agent

  4. Configure the agent’s identity, instructions and tools

  5. Save the agent

Once created, the custom agent becomes available to the users or groups you selected.


Who Can Create and Manage Custom Agents

Custom agents are managed in the Admin section of Moterra.
Only users with admin permissions can create and manage custom agents.


When to Use Custom Agents

Use custom agents when:

  • The General Assistant is too broad for your needs

  • You want more control over assistant behaviour

  • Different teams need different capabilities or access

  • Consistent behaviour across similar tasks is important

For simple or one-off requests, the General Assistant may be sufficient.


What to Read Next