Skip to navigation
Simulation Scenarios

Create Scenario

Creates a scenario, the AI tester's intent plus the success criteria that score a run.

Authentication

AuthorizationBearer

Authenticate server-side requests with a private Vapi API key. Create or copy a key from the Vapi Dashboard and send it in the Authorization header as Bearer <token>. Keep private API keys out of client-side code and public repositories.

Request

This endpoint expects an object.
namestringRequired<=80 characters

The display name of the scenario, for example Book an appointment.

instructionsstringRequired<=10000 characters

What the AI tester should try to accomplish in the conversation. Write it as the AI tester’s goal, for example Book an appointment for next week and confirm the time.

evaluationslist of objectsRequired
The checks that decide whether a run passes. Each evaluation compares a structured output against an expected value. At least one evaluation is required to run.
hookslist of objectsOptional
Hooks to run on simulation lifecycle events
targetOverridesobjectOptional
Overrides to inject into the simulated target assistant or squad
toolMockslist of objectsOptional
Mock results for the assistant or squad's tools during the simulation, so the run stays deterministic without calling real services.
latencyExpectationslist of objectsOptional

Latency ceilings for voice simulations. Each expectation aggregates the target call’s per-turn latencies and fails the simulation (when required) if the aggregated value exceeds its threshold. Skipped for chat simulations and GPT Live targets; on any other voice simulation, a metric that no turn measured fails.

pathstring or nullOptionalformat: "/^[a-zA-Z0-9][a-zA-Z0-9._-]*(?:\/[a-zA-Z0-9][a-zA-Z0-9._-]*){0,2}$/"<=255 characters

Optional folder path for organizing scenarios. Supports up to 3 levels (e.g., “dept/feature/variant”). Maps to GitOps resource folder structure.

Response

idstringformat: "uuid"
This is the unique identifier for the scenario.
orgIdstringformat: "uuid"
This is the unique identifier for the organization this scenario belongs to.
createdAtdatetime

This is the ISO 8601 date-time string of when the scenario was created.

updatedAtdatetime

This is the ISO 8601 date-time string of when the scenario was last updated.

namestring<=80 characters
This is the name of the scenario.
instructionsstring<=10000 characters

This is the script/instructions for the tester to follow during the simulation.

evaluationslist of objects

This is the structured output-based evaluation plan for the simulation. Each item defines a structured output to extract and evaluate against an expected value.

hookslist of objectsOptional
Hooks to run on simulation lifecycle events
targetOverridesobjectOptional
Overrides to inject into the simulated target assistant or squad
toolMockslist of objectsOptional

Scenario-level tool call mocks to use during simulations.

latencyExpectationslist of objectsOptional

Latency ceilings for voice simulations. Each expectation aggregates the target call’s per-turn latencies and fails the simulation (when required) if the aggregated value exceeds its threshold. Skipped for chat simulations and GPT Live targets; on any other voice simulation, a metric that no turn measured fails.

pathstring or nullOptionalformat: "/^[a-zA-Z0-9][a-zA-Z0-9._-]*(?:\/[a-zA-Z0-9][a-zA-Z0-9._-]*){0,2}$/"<=255 characters

Optional folder path for organizing scenarios. Supports up to 3 levels (e.g., “dept/feature/variant”). Maps to GitOps resource folder structure.