Server Configuration

Edit the configuration for Server. Changes require a service restart to take effect.

Configuration JSON Retrieved: 2026-04-18 05:42:42
Server Configuration Reference
SectionKeyDescription
MqttBrokerHostMQTT broker IP/hostname
BrokerPortDefault: 1883
UsernameBroker credentials (optional)
PasswordBroker credentials (optional)
UseTlsEnable TLS for MQTT
HomeAssistantBaseUrlHA instance URL
AccessTokenLong-lived access token
PhoneDeviceTrackerIdEntity ID for proximity checking
HomeZoneEntityIdZone entity for "home" detection
LlmProviderLLM provider: Ollama, LmStudio, or OpenAiCompatible
BaseUrlLLM API base URL (e.g., http://localhost:11434 for Ollama, http://localhost:1234 for LM Studio)
ModelModel name (e.g., llama3.1:8b)
TemperatureLLM creativity (0.0–1.0)
MaxTokensMax response tokens
AudioWakeWordTrigger phrase
UseVoiceActivityDetectionEnable VAD
UseWakeWordRequire wake word
WhisperModelSTT model (large-v3, tiny)
PiperVoiceTTS voice model
SilenceTimeoutMsSilence detection timeout
ApprovalTimeoutSecondsApproval request timeout
MaxPinAttemptsMax PIN attempts before lockout
LockoutMinutesLockout duration
PinHashBCrypt hash of approval PIN
ServerGrpcAddressgRPC listen address
ApiBaseUrlREST API base URL
DatabasePathSQLite database path
AllowedApplicationsApps the Agent can launch (array)
WakeOnLanServerMacAddressServer NIC MAC for WOL
BroadcastAddressNetwork broadcast address
PortWOL port (default: 9)
HealthCheckUrlURL to ping server health
HealthCheckIntervalMsPoll interval
HealthCheckTimeoutMsMax wait time for server