Also known as: agent template, pre-built agent, agent job function
An agent role is a pre-defined job function that shapes how an AI agent behaves: its responsibilities, default instructions, and typical tool permissions. Examples include Project Manager, Account Manager, Payment Recovery Specialist, and Sales Development Rep. Roles are starting points. Each agent created from a role can be customized further.
Where you'll see agent role in day-to-day work inside Helm.
Helm ships with pre-built roles backed by the `agent_roles` table. When you create an agent, you pick a role. That sets the default instructions (what the agent should pay attention to), the recommended tool groups (which apps it can read and write), and the skill set (domain context it uses). You can then customize every part of the configuration per agent: adjust instructions, restrict tools, add custom skills, and set autonomy level. Custom roles can also be authored from scratch.
Concepts that show up in the same workflows and reports.
Common questions and honest answers.
Yes. A common pattern: one Account Manager per client tier (white-glove, standard, long-tail) with different autonomy and cadence settings. Each agent has its own name, memory, and budget.
Yes. Custom roles are first-class. Write the responsibilities, instructions, default tool groups, and the role is available for reuse across the workspace.
Helm is the AI work platform where these concepts stop being theory and start being your Monday morning.