Skip to main content
POST
/
v1
/
accounts
/
{account_id}
/
agents
Create an agent
curl --request POST \
  --url https://api.itellico.ai/v1/accounts/{account_id}/agents \
  --header 'Content-Type: application/json' \
  --header 'X-API-Key: <api-key>' \
  --data '
{
  "model": {
    "max_tokens": 1024,
    "model": "gpt-5-mini",
    "provider": "azure_openai",
    "temperature": 0.7
  },
  "name": "Customer Support Agent",
  "transcriber": {
    "language": "multi",
    "model": "nova-3:general",
    "provider": "deepgram"
  },
  "voice": {
    "provider": "elevenlabs",
    "settings": {
      "similarity_boost": 0.7,
      "stability": 0.7,
      "use_speaker_boost": true
    },
    "voice_id": "pMsXgVXv3BLzUgSXRplE"
  }
}
'
{
  "id": "<string>",
  "account_id": "<string>",
  "name": "<string>",
  "initial_message": {
    "message": "",
    "mode": "fixed_message",
    "delay_ms": 0,
    "interruptible": false
  },
  "ambient_sound": {
    "source": "open_plan_office",
    "volume": 0.5
  },
  "inactivity_settings": {
    "reminder_timeout_ms": 10000,
    "reminder_max_count": 2,
    "end_call_timeout_ms": 305000,
    "reset_on_activity": true
  },
  "response_timing": {
    "min_endpointing_delay_seconds": 0.1
  },
  "max_duration_seconds": 123,
  "note": "<string>",
  "tags": [
    "<string>"
  ],
  "interrupt_settings": {
    "enabled": true,
    "min_words": 0,
    "min_speech_seconds": 0.2
  },
  "model": {},
  "transcriber": {},
  "voice": {},
  "metadata": {},
  "capture_settings": {
    "recording_enabled": true
  },
  "volume": {
    "web": 0.5,
    "telephony": 0.5,
    "allow_adjustment": true
  },
  "denoising": {
    "web": true,
    "telephony": true
  },
  "created": "2023-11-07T05:31:56Z",
  "modified": "2023-11-07T05:31:56Z"
}

Authorizations

​
X-API-Key
string
header
required

Path Parameters

​
account_id
string
required

Body

application/json

Create a new AI agent for handling voice conversations

​
model
OpenAIModelSchema · object
required

OpenAI-specific model configuration.

​
transcriber
AzureTranscriberSchema · object
required

Azure-specific transcriber configuration.

​
voice
AzureVoiceSchema · object
required

Azure-specific voice configuration.

​
name
string | null

The name of the agent. Only used for your own reference to identify and manage agents. Not visible to end users during conversations.

Maximum string length: 50
​
initial_message
InitialMessageSchema · object

Configuration for the agent's initial message when starting a conversation.

​
ambient_sound
AmbientSoundSchema · object

Configuration for ambient background sounds during the conversation.

​
max_duration_seconds
integer | null
default:1200

Maximum allowed length for the conversation in seconds. Default is 1200 seconds (20 minutes) if not specified.

Required range: 10 <= x <= 43200
​
inactivity_settings
InactivitySettings · object

Configuration for handling user inactivity and silence during conversations.

​
note
string | null

Internal notes about the agent. These notes are for your team's reference only and are not visible to end users. Use this to document agent configuration, purpose, or any special instructions.

Maximum string length: 5000
​
tags
string[] | null

List of tags to categorize and organize your agents. Tags help you filter and find agents quickly. Examples: 'sales', 'support', 'lead-qualification', 'appointment-booking'.

​
interrupt_settings
InterruptSettings · object

Configuration for how the agent handles user interruptions during conversation. If not provided, defaults to allowing interruptions with minimal filtering.

​
response_timing
ResponseTimingSchema · object

Configuration for agent response timing and conversation flow control.

​
metadata
Metadata · object

Custom metadata for the agent. Store any additional key-value pairs that your application needs. This data is not used by the agent itself but can be useful for integrations, tracking, or custom business logic.

​
capture_settings
CaptureSettingsSchema · object

Recording configuration for conversation capture. Controls whether conversations are recorded for quality assurance, training, or compliance purposes.

​
volume
VolumeSchema · object

Volume settings for agent audio output. Controls the volume level for different call types and whether users can adjust volume via voice commands.

​
denoising
DenoisingSchema · object

Denoising/noise cancellation settings for enhanced audio quality powered by Krisp. Controls advanced noise suppression for different call types.

Response

201 - application/json

Created

Response returned after agent operations (create, get, update)

​
id
string<uuid4>
required

Unique identifier for the agent. Use this ID to reference the agent in API calls for updates, deletion, or starting conversations.

​
account_id
string<uuid4>
required

Unique identifier for the account that owns this agent.

​
name
string
required

The display name of the agent as configured. This is for your reference and internal organization.

​
initial_message
InitialMessageSchema · object
required

Configuration for the agent's initial message when starting a conversation.

​
ambient_sound
AmbientSoundSchema · object
required

Configuration for ambient background sounds during the conversation.

​
inactivity_settings
InactivitySettings · object
required

Configuration for handling user inactivity and silence during conversations.

​
response_timing
ResponseTimingSchema · object
required

Configuration for agent response timing and conversation flow control.

​
max_duration_seconds
integer | null

The maximum conversation duration configured for this agent in seconds. Maximum allowed is 7200 seconds (2 hours).

​
note
string | null

Internal notes about the agent for your team's reference.

​
tags
string[] | null

List of tags assigned to this agent for categorization and filtering.

​
interrupt_settings
InterruptSettings · object

Configuration for how the agent handles user interruptions during conversation.

​
model
Model · object

Language model configuration for the agent.

​
transcriber
Transcriber · object

Speech-to-text configuration for the agent.

​
voice
Voice · object

Text-to-speech configuration for the agent.

​
metadata
Metadata · object

Custom metadata associated with the agent.

​
capture_settings
CaptureSettingsSchema · object

Capture settings configuration for recording and artifact settings.

​
volume
VolumeSchema · object

Volume settings for different call types and voice adjustment control.

​
denoising
DenoisingSchema · object

Denoising/noise cancellation settings for enhanced audio quality powered by Krisp.

​
created
string<date-time> | null

Date-time of when the agent was created (ISO 8601 on output).

​
modified
string<date-time> | null

Date-time of when the agent was last updated (ISO 8601 on output).