Worker settings
An owner or admin edits a worker in sections: profile, responsibilities, instructions, brain, governance and identities. Each save can carry a reason.
Open a worker on Workers and select Edit. Each section saves on its own. Each save takes an optional Reason for the change, which goes to the audit log.
ProfileLink to Profile
Role, name, handle, team, display name, pronouns, bio, tone, working style and avatar. Handles cannot change.
ResponsibilitiesLink to Responsibilities
Objectives, what is in scope, what is out of scope, success criteria and service expectations.
InstructionsLink to Instructions
Extra instructions for this worker, on top of its role.
BrainLink to Brain
- Model chain: the models for the Fast, Balanced and Deep tiers.
- Reasoning strategy: Plan, act, verify; ReAct; or Reflexion.
- Reasoning effort: Low, Medium or High.
- Max steps per run and Max tool calls per run.
GovernanceLink to Governance
- Daily budget (USD): the most the worker spends in one day. 0 blocks all spend.
- Per-item budget (USD): the most it spends on one work item.
- Concurrent items: how many items it works on at the same time.
- Allowed capabilities and Denied capabilities. A denied capability always wins.
- Escalation contacts.
- Working hours: time zone, start, end and days.
- Autonomy: the worker's own autonomy mode. Only owners and admins can change it.
The Governance tab also shows the organization settings for this worker, with the layer each value comes from.
IdentitiesLink to Identities
The worker's accounts in other tools: GitHub login, GitLab username, Linear user, Jira account, Slack member ID and Teams user ID.
Checked against the product on 2026-10-05.
Was this helpful?
Related articles
- WorkersA worker is an AI employee with a name, a role, a level, a brain and a budget. The Workers page lists them and opens each one.
- Budgets and spend limitsBudgets cap model spend per day for the organization and each worker, and per work item. Agent reserves spend before each call, so it never goes over by more than one call.
- Brain and modelsChoose model defaults, connect providers, set fallback order, inspect activity and choose where work runs.
- Autonomy modesThe autonomy mode decides how much workers do before they ask you. Auto is the default: internal work runs at once, and outside or risky actions wait for you.