Generates flows, metrics and a run plan for the agent. Runs in the background: watch the agent.setup.completed webhook, or poll GET /v1/agent/{agentId}/setup. Refuses while another setup for the agent is running. Needs the ‘simulation:create’ and ‘metric:create’ permissions. Exploration also needs ‘simulation:run’, and is turned off for keys without it unless explore.enabled is set.
roark agent setup create <agent-id> [options]
Calls POST /v1/agent/{agentId}/setup
Arguments
string
required
The agent-id.
Options
string
The industry the agent serves. One of HEALTHCARE, FINANCIAL_SERVICES, INSURANCE, DEBT_COLLECTION, RETAIL, TRAVEL_AND_HOSPITALITY, TELECOMMUNICATIONS, TECHNOLOGY, PUBLIC_SECTOR, UTILITIES, LOGISTICS, AUTOMOTIVE, HOME_SERVICES, RESTAURANTS, EDUCATION, OTHER
string
A page Roark reads to learn about the business.
string
What you want tested, in your own words. Each scenario you describe becomes a flow.
array
The capabilities to test. Omit to let Roark work them out from the agent. One of APPOINTMENT_BOOKING, FAQ, IDENTITY_VERIFICATION, ACCOUNT_MANAGEMENT, BILLING_AND_PAYMENTS, ORDER_MANAGEMENT, TROUBLESHOOTING, CALL_ROUTING, LEAD_QUALIFICATION, COLLECTIONS, OUTBOUND_CALLING Repeatable.
string
The endpoint the run plan calls. Defaults to the agent’s newest endpoint.
object
Exploration settings.
switch
Whether Roark calls the agent to discover what it does before finishing setup. Exploration calls are billed.
integer
The most exploration calls Roark places.
integer
The longest a single exploration call may last, in minutes.
