mv.assistants

AI voice assistants: prompt, voice, tools and call behaviour. These methods are on mv.assistants, where mv is your client (see the Node.js SDK). Each one returns the response object and throws when the API answers with an error.

create()

Create an assistant. Creates an assistant. Only name is required; fields you leave out take their defaults (GET /v1/assistants/{id} shows them all).

mv.assistants.create(body: AssistantsCreateData["body"], options?: RequestOptions): Promise<AssistantsCreateResponse>

POST /assistants · API reference

Parameters

Name	Type	Required	Description
body.name	string	yes	The assistant’s name (yours; callers never hear it).
body.advanced	object	no
body.answering_machine_detection	object	no
body.artifacts	object	no
body.end_call_message	string	no	Said before the assistant ends the call.
body.end_call_phrases	string[]	no	Phrases that end the call.
body.first_message	string	no	What the assistant says first.
body.first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	no
body.flow_id	string | null	no	flow_… ID.
body.language	string	no	The call language.
body.llm	object	no
body.max_duration_seconds	number	no	10–43200 (12 hours).
body.metadata	MetadataInput	no	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
body.public	boolean	no	Allow publishable keys (the browser widget) to call this assistant. Default false.
body.system_prompt	string	no	The instructions the model follows.
body.tools	object	no	Fields you leave out are kept.
body.transcriber	object	no
body.voice	object	no
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options.headers["Idempotency-Key"]	string	no	A unique key (for example a UUID) that makes this request safe to retry: for 24 hours, a retry with the same key and parameters returns the first response instead of acting twice.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

Assistant:

Field	Type	Description
id	string	The assistant’s ID.
object	"assistant"	Always assistant.
livemode	boolean	true in live mode, false in test mode.
name	string	The assistant’s name (yours; callers never hear it).
language	string	The call language (BCP-47, e.g. he, en).
system_prompt	string	The instructions the model follows.
first_message	string	What the assistant says first.
first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	Whether the assistant speaks first (first_message), waits for the caller, or generates its opening.
voice	AssistantVoice	How the assistant sounds.
transcriber	AssistantTranscriber	How the assistant hears.
llm	AssistantLlm	The language model behind the assistant.
tools	AssistantTools	The tools the assistant can use.
flow_id	string | null	flow_… ID.
end_call_message	string	Said before the assistant ends the call.
end_call_phrases	string[]	Phrases that end the call when the assistant says them.
max_duration_seconds	number	Calls end after this long.
answering_machine_detection	AssistantAnsweringMachineDetection	Answering-machine detection on outbound calls.
artifacts	AssistantArtifacts	What is kept after a call.
public	boolean	Publishable keys (the browser widget) may start web calls to this assistant.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
advanced	AssistantAdvanced	Pipeline tuning. The defaults suit most assistants.
created	string	An ISO-8601 timestamp in UTC.
updated	string	An ISO-8601 timestamp in UTC.

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

const assistant = await mv.assistants.create({
  first_message: "שלום, הגעתם למרפאת השרון. איך אפשר לעזור?",
  language: "he",
  metadata: {
    crm_id: "0031x00000AbCdE",
  },
  name: "Clinic receptionist",
  system_prompt: "You book, move and cancel appointments. Keep answers short.",
  tools: {
    custom: [
      {
        description: "Find free appointment slots for a doctor and date",
        headers: {
          Authorization: "Bearer crm-secret",
        },
        name: "find_slots",
        parameters: {
          properties: {
            date: {
              type: "string",
            },
            doctor: {
              type: "string",
            },
          },
          required: [
            "date",
          ],
          type: "object",
        },
        url: "https://crm.example.com/voice/tools",
      },
    ],
    end_call: true,
    transfer_call: {
      destination: "queue:reception",
    },
  },
  voice: {
    provider: "gemini",
    style: "warm",
    voice_id: "Kore",
  },
});
console.log(assistant);

delete()

Delete an assistant. Deletes the assistant. Past calls keep their history; inbound routes and campaigns that used it need another assistant.

mv.assistants.delete(id: AssistantsDeleteData["path"]["id"], options?: RequestOptions): Promise<AssistantsDeleteResponse>

DELETE /assistants/{id} · API reference

Parameters

Name	Type	Required	Description
id	string	yes	A assistant ID (asst_…).
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options.headers["Idempotency-Key"]	string	no	A unique key (for example a UUID) that makes this request safe to retry: for 24 hours, a retry with the same key and parameters returns the first response instead of acting twice.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

DeletedAssistant:

Field	Type	Description
id	string	A assistant ID (prefix asst_).
object	"assistant"	Always assistant.
deleted	true	—

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

const deletedAssistant = await mv.assistants.delete("asst_ONCFICZKnGVk36Wx4t7ADJ");
console.log(deletedAssistant);

duplicate()

Duplicate an assistant. Creates a copy with the same configuration, tools (including their stored headers) and metadata.

mv.assistants.duplicate(id: AssistantsDuplicateData["path"]["id"], body?: AssistantsDuplicateData["body"], options?: RequestOptions): Promise<AssistantsDuplicateResponse>

POST /assistants/{id}/duplicate · API reference

Parameters

Name	Type	Required	Description
id	string	yes	A assistant ID (asst_…).
body.name	string	no	The copy’s name (default: the original’s name + “ (עותק)“).
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options.headers["Idempotency-Key"]	string	no	A unique key (for example a UUID) that makes this request safe to retry: for 24 hours, a retry with the same key and parameters returns the first response instead of acting twice.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

Assistant:

Field	Type	Description
id	string	The assistant’s ID.
object	"assistant"	Always assistant.
livemode	boolean	true in live mode, false in test mode.
name	string	The assistant’s name (yours; callers never hear it).
language	string	The call language (BCP-47, e.g. he, en).
system_prompt	string	The instructions the model follows.
first_message	string	What the assistant says first.
first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	Whether the assistant speaks first (first_message), waits for the caller, or generates its opening.
voice	AssistantVoice	How the assistant sounds.
transcriber	AssistantTranscriber	How the assistant hears.
llm	AssistantLlm	The language model behind the assistant.
tools	AssistantTools	The tools the assistant can use.
flow_id	string | null	flow_… ID.
end_call_message	string	Said before the assistant ends the call.
end_call_phrases	string[]	Phrases that end the call when the assistant says them.
max_duration_seconds	number	Calls end after this long.
answering_machine_detection	AssistantAnsweringMachineDetection	Answering-machine detection on outbound calls.
artifacts	AssistantArtifacts	What is kept after a call.
public	boolean	Publishable keys (the browser widget) may start web calls to this assistant.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
advanced	AssistantAdvanced	Pipeline tuning. The defaults suit most assistants.
created	string	An ISO-8601 timestamp in UTC.
updated	string	An ISO-8601 timestamp in UTC.

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

const assistant = await mv.assistants.duplicate("asst_ONCFICZKnGVk36Wx4t7ADJ", {
  name: "Clinic receptionist (evening)",
});
console.log(assistant);

list()

List assistants. Your assistants, newest first.

mv.assistants.list(query?: NonNullable<AssistantsListData["query"]>, options?: RequestOptions): PagedList<AssistantsListResponse["data"][number], NonNullable<AssistantsListData["query"]>>

GET /assistants · API reference

Parameters

Name	Type	Required	Description
query.limit	integer	no	How many objects to return, 1–100 (default 20).
query.starting_after	string	no	A cursor (next_cursor) or object ID: return the objects after it (older).
query.ending_before	string	no	A cursor or object ID: return the objects before it (newer).
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

A PagedList: await it for the first page, for await it for every item.

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

for await (const assistant of mv.assistants.list()) {
  console.log(assistant);
}

retrieve()

Retrieve an assistant.

mv.assistants.retrieve(id: AssistantsRetrieveData["path"]["id"], options?: RequestOptions): Promise<AssistantsRetrieveResponse>

GET /assistants/{id} · API reference

Parameters

Name	Type	Required	Description
id	string	yes	A assistant ID (asst_…).
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

Assistant:

Field	Type	Description
id	string	The assistant’s ID.
object	"assistant"	Always assistant.
livemode	boolean	true in live mode, false in test mode.
name	string	The assistant’s name (yours; callers never hear it).
language	string	The call language (BCP-47, e.g. he, en).
system_prompt	string	The instructions the model follows.
first_message	string	What the assistant says first.
first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	Whether the assistant speaks first (first_message), waits for the caller, or generates its opening.
voice	AssistantVoice	How the assistant sounds.
transcriber	AssistantTranscriber	How the assistant hears.
llm	AssistantLlm	The language model behind the assistant.
tools	AssistantTools	The tools the assistant can use.
flow_id	string | null	flow_… ID.
end_call_message	string	Said before the assistant ends the call.
end_call_phrases	string[]	Phrases that end the call when the assistant says them.
max_duration_seconds	number	Calls end after this long.
answering_machine_detection	AssistantAnsweringMachineDetection	Answering-machine detection on outbound calls.
artifacts	AssistantArtifacts	What is kept after a call.
public	boolean	Publishable keys (the browser widget) may start web calls to this assistant.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
advanced	AssistantAdvanced	Pipeline tuning. The defaults suit most assistants.
created	string	An ISO-8601 timestamp in UTC.
updated	string	An ISO-8601 timestamp in UTC.

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

const assistant = await mv.assistants.retrieve("asst_ONCFICZKnGVk36Wx4t7ADJ");
console.log(assistant);

update()

Update an assistant. Updates the fields you send. Nested objects merge with the stored values and arrays are replaced; tools.custom replaces the whole list (a tool sent without headers keeps the stored headers of the tool with the same name). Settings that are not part of the API are kept.

mv.assistants.update(id: AssistantsUpdateData["path"]["id"], body?: AssistantsUpdateData["body"], options?: RequestOptions): Promise<AssistantsUpdateResponse>

PATCH /assistants/{id} · API reference

Parameters

Name	Type	Required	Description
id	string	yes	A assistant ID (asst_…).
body.advanced	object	no
body.answering_machine_detection	object	no
body.artifacts	object	no
body.end_call_message	string	no	Said before the assistant ends the call.
body.end_call_phrases	string[]	no	Phrases that end the call.
body.first_message	string	no	What the assistant says first.
body.first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	no
body.flow_id	string | null	no	flow_… ID.
body.language	string	no	The call language.
body.llm	object	no
body.max_duration_seconds	number	no	10–43200 (12 hours).
body.metadata	MetadataInput	no	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
body.name	string	no	The assistant’s name (yours; callers never hear it).
body.public	boolean	no	Allow publishable keys (the browser widget) to call this assistant. Default false.
body.system_prompt	string	no	The instructions the model follows.
body.tools	object	no	Fields you leave out are kept.
body.transcriber	object	no
body.voice	object	no
options.headers["MoreVoice-Version"]	string	no	The API version to use for this request. Defaults to the version the API key is pinned to.
options	RequestOptions	no	idempotencyKey, extra headers and an abort signal: see retries and idempotency.

Returns

Assistant:

Field	Type	Description
id	string	The assistant’s ID.
object	"assistant"	Always assistant.
livemode	boolean	true in live mode, false in test mode.
name	string	The assistant’s name (yours; callers never hear it).
language	string	The call language (BCP-47, e.g. he, en).
system_prompt	string	The instructions the model follows.
first_message	string	What the assistant says first.
first_message_mode	"assistant_speaks_first" | "assistant_waits_for_user" | "assistant_generates_first_message"	Whether the assistant speaks first (first_message), waits for the caller, or generates its opening.
voice	AssistantVoice	How the assistant sounds.
transcriber	AssistantTranscriber	How the assistant hears.
llm	AssistantLlm	The language model behind the assistant.
tools	AssistantTools	The tools the assistant can use.
flow_id	string | null	flow_… ID.
end_call_message	string	Said before the assistant ends the call.
end_call_phrases	string[]	Phrases that end the call when the assistant says them.
max_duration_seconds	number	Calls end after this long.
answering_machine_detection	AssistantAnsweringMachineDetection	Answering-machine detection on outbound calls.
artifacts	AssistantArtifacts	What is kept after a call.
public	boolean	Publishable keys (the browser widget) may start web calls to this assistant.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
advanced	AssistantAdvanced	Pipeline tuning. The defaults suit most assistants.
created	string	An ISO-8601 timestamp in UTC.
updated	string	An ISO-8601 timestamp in UTC.

Example

import { createClient, createResources } from "@morevoice/sdk";

const mv = createResources({
  client: createClient({ baseUrl: "https://api.morevoice.ai/v1", auth: process.env.MOREVOICE_API_KEY }),
});

const assistant = await mv.assistants.update("asst_ONCFICZKnGVk36Wx4t7ADJ", {
  advanced: {
    stop_speaking_plan: {
      num_words: 2,
    },
  },
  metadata: {
    stage: "beta",
  },
  voice: {
    style: "calm and reassuring",
  },
});
console.log(assistant);