> ## Documentation Index > Fetch the complete documentation index at: https://developers.telnyx.com/llms.txt > Use this file to discover all available pages before exploring further. # List assistants > Retrieve a list of all AI Assistants configured by the user. ## OpenAPI ````yaml https://telnyx-openapi-ng.s3.us-east-1.amazonaws.com/ai/assistants.yml get /ai/assistants openapi: 3.1.0 info: title: Telnyx AI Assistants API version: 2.0.0 description: >- API for managing AI Assistants, including CRUD fields, versions, tags, integrations, MCP servers, and shared tools. contact: email: support@telnyx.com servers: - url: https://api.telnyx.com/v2 security: - bearerAuth: [] paths: /ai/assistants: get: tags: - Assistants summary: List assistants description: Retrieve a list of all AI Assistants configured by the user. operationId: get_assistants_public_assistants_get responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/AssistantsListData' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' x-codeSamples: - lang: JavaScript source: |- import Telnyx from 'telnyx'; const client = new Telnyx({ apiKey: process.env['TELNYX_API_KEY'], // This is the default and can be omitted }); const assistantsList = await client.ai.assistants.list(); console.log(assistantsList.data); - lang: Python source: |- import os from telnyx import Telnyx client = Telnyx( api_key=os.environ.get("TELNYX_API_KEY"), # This is the default and can be omitted ) assistants_list = client.ai.assistants.list() print(assistants_list.data) - lang: Go source: "package main\n\nimport (\n\t\"context\"\n\t\"fmt\"\n\n\t\"github.com/team-telnyx/telnyx-go\"\n\t\"github.com/team-telnyx/telnyx-go/option\"\n)\n\nfunc main() {\n\tclient := telnyx.NewClient(\n\t\toption.WithAPIKey(\"My API Key\"),\n\t)\n\tassistantsList, err := client.AI.Assistants.List(context.TODO())\n\tif err != nil {\n\t\tpanic(err.Error())\n\t}\n\tfmt.Printf(\"%+v\\n\", assistantsList.Data)\n}\n" - lang: Java source: |- package com.telnyx.sdk.example; import com.telnyx.sdk.client.TelnyxClient; import com.telnyx.sdk.client.okhttp.TelnyxOkHttpClient; import com.telnyx.sdk.models.ai.assistants.AssistantListParams; import com.telnyx.sdk.models.ai.assistants.AssistantsList; public final class Main { private Main() {} public static void main(String[] args) { TelnyxClient client = TelnyxOkHttpClient.fromEnv(); AssistantsList assistantsList = client.ai().assistants().list(); } } - lang: Ruby source: |- require "telnyx" telnyx = Telnyx::Client.new(api_key: "My API Key") assistants_list = telnyx.ai.assistants.list puts(assistants_list) - lang: PHP source: >- ai->assistants->list(); var_dump($assistantsList); } catch (APIException $e) { echo $e->getMessage(); } - lang: CLI source: |- telnyx ai:assistants list \ --api-key 'My API Key' components: schemas: AssistantsListData: properties: data: items: $ref: '#/components/schemas/inference-embedding_Assistant' type: array title: Data type: object required: - data title: AssistantsListData HTTPValidationError: title: HTTPValidationError type: object properties: detail: title: Detail type: array items: $ref: '#/components/schemas/ValidationError' inference-embedding_Assistant: properties: id: type: string name: type: string created_at: type: string format: date-time version_id: type: string description: >- Identifier for the assistant version returned by version-aware assistant endpoints. version_created_at: type: string format: date-time description: Timestamp when this assistant version was created. description: type: string model: type: string description: >- ID of the model to use when `external_llm` is not set. You can use the [Get models API](https://developers.telnyx.com/api-reference/openai-chat/get-available-models-openai-compatible) to see available models. If `external_llm` is provided, the assistant uses `external_llm` instead of this field. If neither `model` nor `external_llm` is provided, Telnyx applies the default model. instructions: type: string description: >- System instructions for the assistant. These may be templated with [dynamic variables](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables) tools: $ref: '#/components/schemas/AssistantTools' description: >- Deprecated for new integrations. Inline tool definitions available to the assistant. Prefer `tool_ids` to attach shared tools created with the AI Tools endpoints. mcp_servers: type: array items: $ref: '#/components/schemas/AssistantMCPServer' default: [] description: >- MCP servers attached to the assistant. Create MCP servers with `/ai/mcp_servers`, then reference them by `id` here. greeting: type: string description: >- Text that the assistant will use to start the conversation. This may be templated with [dynamic variables](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables). Use an empty string to have the assistant wait for the user to speak first. Use the special value `` to have the assistant generate the greeting based on the system instructions. llm_api_key_ref: type: string description: >- This is only needed when using third-party inference providers selected by `model`. The `identifier` for an integration secret [/v2/integration_secrets](https://developers.telnyx.com/api-reference/integration-secrets/create-a-secret) that refers to your LLM provider's API key. For bring-your-own endpoint authentication, use `external_llm.llm_api_key_ref` instead. Warning: Free plans are unlikely to work with this integration. external_llm: $ref: '#/components/schemas/ExternalLLM' fallback_config: $ref: '#/components/schemas/FallbackConfig' voice_settings: $ref: '#/components/schemas/VoiceSettings' transcription: $ref: '#/components/schemas/TranscriptionSettings' telephony_settings: $ref: '#/components/schemas/TelephonySettings' messaging_settings: $ref: '#/components/schemas/MessagingSettings' enabled_features: items: $ref: '#/components/schemas/EnabledFeatures' type: array uniqueItems: true title: Enabled Features insight_settings: $ref: '#/components/schemas/InsightSettings' privacy_settings: $ref: '#/components/schemas/PrivacySettings' dynamic_variables_webhook_url: type: string description: >- If `dynamic_variables_webhook_url` is set, Telnyx sends a POST request to this URL at the start of the conversation to resolve dynamic variables. **Gotcha:** the webhook response must wrap variables under a top-level `dynamic_variables` object, e.g. `{"dynamic_variables": {"customer_name": "Jane"}}`. Returning a flat object will be ignored and variables will fall back to their defaults. See the [dynamic variables guide](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables) for the full request/response format and timeout behavior. dynamic_variables_webhook_timeout_ms: type: integer minimum: 1 maximum: 10000 default: 1500 description: >- Timeout in milliseconds for the dynamic variables webhook. Must be between 1 and 10000 ms. If the webhook does not respond within this timeout, the call proceeds with default values. See the [dynamic variables guide](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables). dynamic_variables: type: object description: Map of dynamic variables and their values additionalProperties: true import_metadata: $ref: '#/components/schemas/ImportMetadata' widget_settings: $ref: '#/components/schemas/WidgetSettings' interruption_settings: $ref: '#/components/schemas/inference-embedding_InterruptionSettings' integrations: type: array items: $ref: '#/components/schemas/AssistantIntegration' default: [] description: >- Connected integrations attached to the assistant. The catalog of available integrations is at `/ai/integrations`; the user's connected integrations are at `/ai/integrations/connections`. Each item references a catalog integration by `integration_id`. observability_settings: $ref: '#/components/schemas/Observability' version_name: type: string maxLength: 50 default: New assistant description: Human-readable name for the assistant version. related_mission_ids: type: array items: type: string default: [] description: IDs of missions related to this assistant. tags: type: array items: type: string default: [] description: >- Tags associated with the assistant. Tags can also be managed with the assistant tag endpoints. post_conversation_settings: $ref: '#/components/schemas/PostConversationSettings' conversation_flow: $ref: '#/components/schemas/ConversationFlow' type: object required: - id - name - created_at - model - instructions ValidationError: title: ValidationError required: - loc - msg - type type: object properties: loc: title: Location type: array items: anyOf: - type: string - type: integer msg: title: Message type: string type: title: Error Type type: string AssistantTools: description: >- Deprecated for new integrations. Inline tool definitions available to the assistant. Prefer `tool_ids` to attach shared tools created with the AI Tools endpoints. type: array items: oneOf: - $ref: '#/components/schemas/inference-embedding_WebhookTool' - $ref: '#/components/schemas/RetrievalTool' - $ref: '#/components/schemas/HandoffTool' - $ref: '#/components/schemas/HangupTool' - $ref: '#/components/schemas/inference-embedding_TransferTool' - $ref: '#/components/schemas/InviteTool' - $ref: '#/components/schemas/SIPReferTool' - $ref: '#/components/schemas/DTMFTool' - $ref: '#/components/schemas/SendMessageTool' - $ref: '#/components/schemas/SkipTurnTool' AssistantMCPServer: type: object title: AssistantMCPServer description: >- Reference to an MCP server attached to an assistant. Create and manage MCP servers with the `/ai/mcp_servers` endpoints, then attach them to assistants by ID. properties: id: type: string description: >- ID of the MCP server to attach. This must be the `id` of an MCP server returned by the `/ai/mcp_servers` endpoints. allowed_tools: type: array items: type: string description: >- Optional per-assistant allowlist of MCP tool names. When omitted, the assistant uses the MCP server's configured `allowed_tools`. required: - id ExternalLLM: properties: model: type: string description: Model identifier to use with the external LLM endpoint. base_url: type: string description: Base URL for the external LLM endpoint. llm_api_key_ref: type: string description: Integration secret identifier for the external LLM API key. authentication_method: $ref: '#/components/schemas/AuthenticationMethod' certificate_ref: type: string description: >- Integration secret identifier for the client certificate used with certificate authentication. token_retrieval_url: type: string description: >- URL used to retrieve an access token when certificate authentication is enabled. forward_metadata: type: boolean default: false description: >- When `true`, Telnyx forwards the assistant's dynamic variables to the external LLM endpoint as a top-level `extra_metadata` object on the chat completion request body. Defaults to `false`. Example payload sent to the external endpoint: `{"extra_metadata": {"customer_name": "Jane", "account_id": "acct_789", "telnyx_agent_target": "+13125550100", "telnyx_end_user_target": "+13125550123"}}`. Distinct from OpenAI's native `metadata` field, which has its own size and type limits. type: object required: - model - base_url title: ExternalLLM FallbackConfig: properties: model: type: string description: >- Fallback Telnyx-hosted model to use when the primary LLM provider is unavailable. llm_api_key_ref: type: string description: Integration secret identifier for the fallback model API key. external_llm: $ref: '#/components/schemas/ExternalLLM' type: object title: FallbackConfig VoiceSettings: properties: voice: type: string description: >- The voice to be used by the voice assistant. Check the full list of [available voices](https://developers.telnyx.com/docs/tts-stt/tts-available-voices) via our voices API. To use ElevenLabs, you must reference your ElevenLabs API key as an integration secret under the `api_key_ref` field. See [integration secrets documentation](https://developers.telnyx.com/api-reference/integration-secrets/create-a-secret) for details. For Telnyx voices, use `Telnyx..` (e.g. Telnyx.KokoroTTS.af_heart). The voice portion of the identifier supports [dynamic variables](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables) using mustache syntax (e.g. `Telnyx.Ultra.{{voice_id}}`). The variable is resolved at call time from your dynamic variables webhook, allowing you to select the voice dynamically per call. voice_speed: type: number default: 1 description: >- The speed of the voice in the range [0.25, 2.0]. 1.0 is deafult speed. Larger numbers make the voice faster, smaller numbers make it slower. This is only applicable for Telnyx Natural voices. api_key_ref: type: string description: >- The `identifier` for an integration secret [/v2/integration_secrets](https://developers.telnyx.com/api-reference/integration-secrets/create-a-secret) that refers to your ElevenLabs API key. Warning: Free plans are unlikely to work with this integration. temperature: type: number default: 0.5 description: >- Determines how stable the voice is and the randomness between each generation. Lower values create a broader emotional range; higher values produce more consistent, monotonous output. Only applicable when using ElevenLabs. similarity_boost: type: number default: 0.75 description: >- Determines how closely the AI should adhere to the original voice when attempting to replicate it. Only applicable when using ElevenLabs. use_speaker_boost: type: boolean default: true description: >- Amplifies similarity to the original speaker voice. Increases computational load and latency slightly. Only applicable when using ElevenLabs. style: type: number default: 0 description: >- Determines the style exaggeration of the voice. Amplifies speaker style but consumes additional resources when set above 0. Only applicable when using ElevenLabs. speed: type: number default: 1 description: >- Adjusts speech velocity. 1.0 is default speed; values less than 1.0 slow speech; values greater than 1.0 accelerate it. Only applicable when using ElevenLabs. language_boost: type: - string - 'null' description: >- Enhances recognition for specific languages and dialects during MiniMax TTS synthesis. Default is null (no boost). Set to 'auto' for automatic language detection. Only applicable when using MiniMax voices. enum: - null - auto - Chinese - Chinese,Yue - English - Arabic - Russian - Spanish - French - Portuguese - German - Turkish - Dutch - Ukrainian - Vietnamese - Indonesian - Japanese - Italian - Korean - Thai - Polish - Romanian - Greek - Czech - Finnish - Hindi - Bulgarian - Danish - Hebrew - Malay - Persian - Slovak - Swedish - Croatian - Filipino - Hungarian - Norwegian - Slovenian - Catalan - Nynorsk - Tamil - Afrikaans default: null expressive_mode: type: boolean default: false description: >- Enables emotionally expressive speech using SSML emotion tags. When enabled, the assistant uses audio tags like angry, excited, content, and sad to add emotional nuance. Only supported for Telnyx Ultra voices. background_audio: description: >- Optional background audio to play on the call. Use a predefined media bed, or supply a looped MP3 URL. If a media URL is chosen in the portal, customers can preview it before saving. oneOf: - type: object properties: type: type: string enum: - predefined_media description: Select from predefined media options. value: type: string enum: - silence - office default: silence description: >- The predefined media to use. `silence` disables background audio. volume: type: number minimum: 0.1 maximum: 1 multipleOf: 0.1 default: 1 description: >- Volume level for the predefined background audio. Supports values from 0.1 to 1.0 in 0.1 increments. required: - type - value - type: object properties: type: type: string enum: - media_url description: >- Provide a direct URL to an MP3 file. The audio will loop during the call. value: type: string format: uri description: HTTPS URL to an MP3 file. required: - type - value - type: object properties: type: type: string enum: - media_name description: >- Reference a previously uploaded media by its name from Telnyx Media Storage. value: type: string description: >- The `name` of a media asset created via [Media Storage API](https://developers.telnyx.com/api/media-storage/create-media-storage). The audio will loop during the call. required: - type - value type: object required: - voice TranscriptionSettings: properties: model: type: string enum: - deepgram/flux - deepgram/nova-3 - deepgram/nova-2 - azure/fast - assemblyai/universal-streaming - xai/grok-stt - soniox/stt-rt-v4 - distil-whisper/distil-large-v2 - openai/whisper-large-v3-turbo description: >- The speech to text model to be used by the voice assistant. All Deepgram models are run on-premise. - `deepgram/flux` is optimized for turn-taking with multilingual language hints. - `deepgram/nova-3` is multilingual with automatic language detection. - `deepgram/nova-2` is Deepgram's previous-generation multilingual model. - `azure/fast` is a multilingual Azure transcription model. - `assemblyai/universal-streaming` is a multilingual streaming model with configurable turn detection. - `xai/grok-stt` is a multilingual Grok STT model. - `soniox/stt-rt-v4` is a multilingual streaming model with automatic language detection and configurable endpointing. language: type: string description: >- The language of the audio to be transcribed. If not set, or if set to `auto`, supported models will automatically detect the language. For `deepgram/flux`, supported values are: `auto` (Telnyx language detection controls the language hint), `multi` (no language hint), and language-specific hints `en`, `es`, `fr`, `de`, `hi`, `ru`, `pt`, `ja`, `it`, and `nl`. For `soniox/stt-rt-v4`, `auto` omits the language hint and lets Soniox auto-detect; ISO 639-1 codes (e.g. `en`, `es`) bias detection toward that language. api_key_ref: title: Api Key Ref type: string description: >- Integration secret identifier for the transcription provider API key. Currently used for Azure transcription regions that require a customer-provided API key. region: title: Region type: string description: >- Region on third party cloud providers (currently Azure) if using one of their models. Some regions require `api_key_ref`. settings: $ref: '#/components/schemas/TranscriptionSettingsConfig' type: object TelephonySettings: properties: default_texml_app_id: type: string description: >- Default Texml App used for voice calls with your assistant. This will be created automatically on assistant creation. supports_unauthenticated_web_calls: type: boolean description: >- When enabled, allows users to interact with your AI assistant directly from your website without requiring authentication. This is required for FE widgets that work with assistants that have telephony enabled. noise_suppression: type: string enum: - krisp - deepfilternet - disabled description: >- The noise suppression engine to use. Use 'disabled' to turn off noise suppression. noise_suppression_config: type: object description: >- Configuration for noise suppression. Only applicable when noise_suppression is 'deepfilternet'. properties: attenuation_limit: type: integer minimum: 0 maximum: 100 default: 100 description: 'Attenuation limit for noise suppression. Range: 0-100.' mode: type: string enum: - advanced description: Mode for noise suppression configuration. time_limit_secs: type: integer minimum: 30 maximum: 14400 default: 1800 description: >- Maximum duration in seconds for the AI assistant to participate on the call. When this limit is reached the assistant will be stopped. This limit does not apply to portions of a call without an active assistant (for instance, a call transferred to a human representative). user_idle_timeout_secs: type: integer minimum: 10 maximum: 14400 description: >- Maximum duration in seconds of end user silence on the call. When this limit is reached the assistant will be stopped. This limit does not apply to portions of a call without an active assistant (for instance, a call transferred to a human representative). user_idle_reply_secs: type: integer minimum: 0 default: 10 description: >- Duration in seconds of end user silence before the assistant checks in on the user. When this limit is reached the assistant will prompt the user to respond. This is distinct from user_idle_timeout_secs which stops the assistant entirely. voicemail_detection: type: object description: >- Configuration for voicemail detection (AMD - Answering Machine Detection) on outgoing calls. These settings only apply if AMD is enabled on the Dial command. See [TeXML Dial documentation](https://developers.telnyx.com/api-reference/texml-rest-commands/initiate-an-outbound-call) for enabling AMD. Recommended settings: MachineDetection=Enable, AsyncAmd=true, DetectionMode=Premium. properties: on_voicemail_detected: type: object description: Action to take when voicemail is detected. properties: action: type: string enum: - stop_assistant - leave_message_and_stop_assistant - continue_assistant description: The action to take when voicemail is detected. voicemail_message: type: object description: >- Configuration for the voicemail message to leave. Only applicable when action is 'leave_message_and_stop_assistant'. properties: type: type: string enum: - prompt - message description: >- The type of voicemail message. Use 'prompt' to have the assistant generate a message based on a prompt, or 'message' to leave a specific message. prompt: type: string description: >- The prompt to use for generating the voicemail message. Only applicable when type is 'prompt'. message: type: string description: >- The specific message to leave as voicemail. Only applicable when type is 'message'. recording_settings: type: object description: Configuration for call recording format and channel settings. properties: enabled: type: boolean default: true description: >- Whether call recording is enabled. When set to false, calls will not be recorded regardless of other recording configuration. channels: type: string enum: - single - dual default: dual description: >- The number of channels for the recording. 'single' for mono, 'dual' for stereo. format: type: string enum: - wav - mp3 default: mp3 description: The format of the recording file. stop_on_conversation_end: type: boolean default: false description: >- When enabled, the call recording will stop when the conversation ends (for example, when the assistant hangs up or the call is transferred). When disabled, recording continues until the call itself ends. type: object title: TelephonySettings MessagingSettings: properties: default_messaging_profile_id: type: string description: >- Default Messaging Profile used for messaging exchanges with your assistant. This will be created automatically on assistant creation. delivery_status_webhook_url: type: string description: >- The URL where webhooks related to delivery statused for assistant messages will be sent. conversation_inactivity_minutes: type: integer minimum: 1 maximum: 10000000 description: >- If more than this many minutes have passed since the last message, the assistant will start a new conversation instead of continuing the existing one. type: object EnabledFeatures: type: string enum: - telephony - messaging description: >- If `telephony` is enabled, the assistant will be able to make and receive calls. If `messaging` is enabled, the assistant will be able to send and receive messages. InsightSettings: properties: insight_group_id: type: string description: >- Reference to an Insight Group. Insights in this group will be run automatically for all the assistant's conversations. type: object PrivacySettings: properties: data_retention: type: boolean description: >- If true, conversation history and insights will be stored. If false, they will not be stored. This in‑tool toggle governs solely the retention of conversation history and insights via the AI assistant. It has no effect on any separate recording, transcription, or storage configuration that you have set at the account, number, or application level. All such external settings remain in force regardless of your selection here. type: object ImportMetadata: properties: import_provider: type: string enum: - elevenlabs - vapi - retell description: Provider the assistant was imported from. import_id: type: string description: ID of the assistant in the provider's system. type: object WidgetSettings: type: object properties: theme: type: string enum: - light - dark description: The visual theme for the widget. audio_visualizer_config: $ref: '#/components/schemas/AudioVisualizerConfig' start_call_text: type: string description: Custom text displayed on the start call button. default_state: type: string enum: - expanded - collapsed description: The default state of the widget. position: type: string enum: - fixed - static description: The positioning style for the widget. view_history_url: type: - string - 'null' description: URL to view conversation history. report_issue_url: type: - string - 'null' description: URL for users to report issues. give_feedback_url: type: - string - 'null' description: URL for users to give feedback. agent_thinking_text: type: string description: Text displayed while the agent is processing. speak_to_interrupt_text: type: string description: Text prompting users to speak to interrupt. logo_icon_url: type: - string - 'null' description: URL to a custom logo icon for the widget. title: WidgetSettings description: Configuration settings for the assistant's web widget. inference-embedding_InterruptionSettings: type: object title: InterruptionSettings description: >- Settings for interruptions and how the assistant decides the user has finished speaking. These timings are most relevant when using non turn-taking transcription models. For turn-taking models like `deepgram/flux`, end-of-turn behavior is controlled by the transcription end-of-turn settings under `transcription.settings` (`eot_threshold`, `eot_timeout_ms`, `eager_eot_threshold`). properties: enable: type: boolean default: true description: Whether users can interrupt the assistant while it is speaking. disable_greeting_interruption: type: boolean description: >- When true, disables user interruptions while the assistant greeting is playing. start_speaking_plan: $ref: '#/components/schemas/StartSpeakingPlan' AssistantIntegration: type: object title: AssistantIntegration description: >- Reference to a connected integration attached to an assistant. Discover available integrations with `/ai/integrations` and connected integrations with `/ai/integrations/connections`. properties: integration_id: type: string description: >- Catalog integration ID to attach. This is the `id` from the integrations catalog at `/ai/integrations` (the same value also appears as `integration_id` on entries returned by `/ai/integrations/connections`). It is **not** the connection-level `id` from `/ai/integrations/connections`. allowed_list: type: array items: type: string description: >- Optional per-assistant allowlist of integration tool names. When omitted or empty, all tools allowed by the connected integration are available to the assistant. required: - integration_id Observability: properties: status: $ref: '#/components/schemas/ObservabilityStatus' default: disabled secret_key_ref: title: Secret Key Ref type: string public_key_ref: title: Public Key Ref type: string host: title: Host type: string prompt_name: title: Prompt Name type: string prompt_version: title: Prompt Version type: integer minimum: 1 prompt_label: title: Prompt Label type: string prompt_sync: $ref: '#/components/schemas/PromptSyncStatus' default: disabled type: object title: Observability PostConversationSettings: type: object description: >- Configuration for post-conversation processing. When enabled, the assistant receives one additional LLM turn after the conversation ends, allowing it to execute tool calls such as logging to a CRM or sending a summary. The assistant can execute multiple parallel or sequential tools during this phase. Telephony-control tools (e.g. hangup, transfer) are unavailable post-conversation. Beta feature. properties: enabled: type: boolean description: >- Whether post-conversation processing is enabled. When true, the assistant will be invoked after the conversation ends to perform any final tool calls. Defaults to false. default: false ConversationFlow: description: Conversation flow as returned by the API. example: edges: - condition: prompt: The caller is asking about a bill or charge. type: llm id: e_intake_to_billing start_node_id: n_intake target: node_id: n_billing type: node - condition: prompt: The caller has explicitly asked for a human. type: llm id: e_intake_to_escalation_assistant start_node_id: n_intake target: assistant_id: assistant-human-handoff position: x: 600 'y': 80 type: assistant voice_mode: distinct nodes: - type: prompt id: n_intake instructions: Greet the caller and ask what they're calling about. name: Intake position: x: 120 'y': 80 shared_tool_ids: - tool-faq-kb - type: prompt id: n_billing instructions: >- Focus on billing questions. Look up the caller's latest invoice with the billing tool before answering. instructions_mode: append model: moonshotai/Kimi-K2.6 name: Billing position: x: 420 'y': 80 shared_tool_ids: - tool-billing-lookup tools_mode: append start_node_id: n_intake properties: edges: description: Directed transitions between nodes. items: $ref: '#/components/schemas/FlowEdge' title: Edges type: array nodes: description: All nodes in the flow. items: discriminator: mapping: prompt: $ref: '#/components/schemas/FlowNode' speak: $ref: '#/components/schemas/SpeakNode' tool: $ref: '#/components/schemas/ToolNode' propertyName: type oneOf: - $ref: '#/components/schemas/FlowNode' - $ref: '#/components/schemas/ToolNode' - $ref: '#/components/schemas/SpeakNode' title: Nodes type: array start_node_id: description: ID of the node where the conversation begins. title: Start Node Id type: string required: - start_node_id - nodes title: ConversationFlow type: object inference-embedding_WebhookTool: properties: type: type: string enum: - webhook webhook: $ref: '#/components/schemas/WebhookToolParams' type: object required: - type - webhook title: WebhookTool RetrievalTool: properties: type: type: string enum: - retrieval retrieval: $ref: '#/components/schemas/BucketIds' type: object required: - type - retrieval title: RetrievalTool HandoffTool: description: >- The handoff tool allows the assistant to hand off control of the conversation to another AI assistant. By default, this will happen transparently to the end user. properties: type: type: string enum: - handoff handoff: $ref: '#/components/schemas/HandoffToolParams' type: object required: - type - handoff title: HandoffTool HangupTool: properties: type: type: string enum: - hangup hangup: $ref: '#/components/schemas/HangupToolParams' type: object required: - type - hangup title: HangupTool inference-embedding_TransferTool: properties: type: type: string enum: - transfer transfer: $ref: '#/components/schemas/InferenceEmbeddingTransferToolParams' type: object required: - type - transfer title: TransferTool InviteTool: properties: type: type: string enum: - invite invite: $ref: '#/components/schemas/InviteToolConfig' title: InviteToolConfig type: object required: - type - invite title: InviteTool SIPReferTool: properties: type: type: string enum: - refer refer: $ref: '#/components/schemas/SIPReferToolParams' type: object required: - type - refer title: SIPReferTool DTMFTool: properties: type: type: string enum: - send_dtmf send_dtmf: type: object additionalProperties: true type: object required: - type - send_dtmf title: DTMFTool SendMessageTool: description: >- The send_message tool allows the assistant to send SMS or MMS messages to the end user. The 'to' and 'from' addresses are automatically determined from the conversation context, and the message text is generated by the assistant unless a message_template is provided for runtime variable substitution. properties: type: type: string enum: - send_message send_message: type: object properties: message_template: type: - string - 'null' description: >- Optional message template with dynamic variable support using mustache syntax (e.g., {{variable_name}}). When set, the assistant will use this template for the SMS body instead of generating one. Dynamic variables like {{telnyx_end_user_target}}, {{telnyx_agent_target}}, and custom webhook-provided variables will be resolved at runtime. additionalProperties: true type: object required: - type - send_message title: SendMessageTool SkipTurnTool: properties: type: type: string enum: - skip_turn skip_turn: $ref: '#/components/schemas/SkipTurnToolParams' type: object required: - type - skip_turn title: SkipTurnTool AuthenticationMethod: type: string enum: - token - certificate default: token description: Authentication method used when connecting to the external LLM endpoint. title: AuthenticationMethod TranscriptionSettingsConfig: properties: smart_format: title: Smart Format type: boolean numerals: title: Numerals type: boolean eot_threshold: title: Eot Threshold type: number description: >- Available only for deepgram/flux. Confidence required to trigger an end of turn. Higher values = more reliable turn detection but slightly increased latency. minimum: 0.5 maximum: 0.9 default: 0.8 eot_timeout_ms: title: Eot Timeout Ms type: integer description: >- Available only for deepgram/flux. Maximum milliseconds of silence before forcing an end of turn, regardless of confidence. minimum: 500 maximum: 10000 default: 5000 eager_eot_threshold: title: Eager Eot Threshold type: number minimum: 0.3 maximum: 0.9 default: 0.8 description: >- Available only for deepgram/flux. Confidence threshold for eager end of turn detection. Must be lower than or equal to eot_threshold. Setting this equal to eot_threshold effectively disables eager end of turn. keyterm: title: Keyterm type: string description: >- Available only for deepgram/nova-3 and deepgram/flux. A comma-separated list of key terms to boost for recognition during transcription. Helps improve accuracy for domain-specific terminology, proper nouns, or uncommon words. This field may be templated with [dynamic variables](https://developers.telnyx.com/docs/inference/ai-assistants/dynamic-variables) using mustache syntax (e.g. `Telnyx,{{customer_name}},VoIP`). Variables are resolved at call time before the value is sent to the speech-to-text engine. end_of_turn_confidence_threshold: title: End Of Turn Confidence Threshold type: number minimum: 0 maximum: 1 default: 0.4 description: >- Available only for assemblyai/universal-streaming. Confidence level required to trigger an end of turn. Higher values require more certainty before ending a turn. min_turn_silence: title: Min Turn Silence type: integer minimum: 100 maximum: 5000 default: 400 description: >- Available only for assemblyai/universal-streaming. Minimum duration of silence in milliseconds before a turn can end. Must be less than or equal to max_turn_silence. max_turn_silence: title: Max Turn Silence type: integer minimum: 100 maximum: 5000 default: 1280 description: >- Available only for assemblyai/universal-streaming. Maximum duration of silence in milliseconds before forcing an end of turn. interim_results: title: Interim Results type: boolean default: false description: >- Available only for soniox/stt-rt-v4. When true, Soniox streams interim (non-final) results in addition to finalized transcripts. enable_endpoint_detection: title: Enable Endpoint Detection type: boolean default: false description: >- Available only for soniox/stt-rt-v4. When true, Soniox emits end-of-utterance events at the cadence configured by `max_endpoint_delay_ms`. max_endpoint_delay_ms: title: Max Endpoint Delay Ms type: integer minimum: 500 maximum: 3000 description: >- Available only for soniox/stt-rt-v4. Maximum silence (in milliseconds) before Soniox emits an end-of-utterance event. Only honored when `enable_endpoint_detection` is true. type: object title: TranscriptionSettingsConfig AudioVisualizerConfig: type: object properties: color: type: string enum: - verdant - twilight - bloom - mystic - flare - glacier description: The color theme for the audio visualizer. preset: type: string description: The preset style for the audio visualizer. title: AudioVisualizerConfig StartSpeakingPlan: type: object title: StartSpeakingPlan description: >- Controls when the assistant starts speaking after the user stops. These thresholds primarily apply to non turn-taking transcription models. For turn-taking models like `deepgram/flux`, end-of-turn detection is driven by the transcription end-of-turn settings under `transcription.settings` instead. properties: wait_seconds: type: number format: float default: 0.4 minimum: 0 description: Minimum seconds to wait before the assistant starts speaking. transcription_endpointing_plan: $ref: '#/components/schemas/TranscriptionEndpointingPlan' ObservabilityStatus: type: string enum: - enabled - disabled title: ObservabilityStatus PromptSyncStatus: type: string enum: - enabled - disabled title: PromptSyncStatus description: >- Whether to auto-publish the assistant's instructions as a Langfuse prompt. When ENABLED + prompt_name set, every assistant create/update pushes `instructions` to Langfuse via create_prompt and stores the returned version in prompt_version. FlowEdge: description: |- Directed transition from one node to a target, gated by a condition. The target is either another node in the same flow (`NodeTarget`) or a different assistant (`AssistantTarget`). Multiple edges may share a `start_node_id`; the runtime evaluates them in the order they're declared and takes the first whose condition is true. properties: condition: description: >- Condition that gates the transition. Discriminated by `type`: `llm`, `expression`. discriminator: mapping: expression: $ref: '#/components/schemas/ExpressionCondition' llm: $ref: '#/components/schemas/LLMCondition' default: $ref: '#/components/schemas/DefaultCondition' propertyName: type oneOf: - $ref: '#/components/schemas/LLMCondition' - $ref: '#/components/schemas/ExpressionCondition' - $ref: '#/components/schemas/DefaultCondition' title: Condition id: description: Caller-supplied unique identifier for this edge within the flow. example: e_age_gate title: Id type: string start_node_id: description: ID of the node this edge transitions away from. title: Start Node Id type: string target: description: >- Destination of the transition. Discriminated by `type`: `node` (jump to another node in this flow) or `assistant` (hand off to a different assistant). discriminator: mapping: assistant: $ref: '#/components/schemas/AssistantTarget' node: $ref: '#/components/schemas/NodeTarget' propertyName: type oneOf: - $ref: '#/components/schemas/NodeTarget' - $ref: '#/components/schemas/AssistantTarget' title: Target required: - id - start_node_id - target - condition title: FlowEdge type: object FlowNode: description: One step in a conversation flow, as returned by the API. properties: external_llm: $ref: '#/components/schemas/ExternalLLM' description: >- Override for `Assistant.external_llm` while this node is active. Use this to route a node's turns to a different external LLM (different `model`, `base_url`, credentials). Part of the LLM bundle — see `model` for cascade semantics. Mutually exclusive with `model` on the node (a single LLM identity per node). id: description: Caller-supplied unique identifier for this node within the flow. title: Id type: string instructions: description: Prompt that drives the LLM while this node is active. Required. title: Instructions type: string instructions_mode: default: replace description: >- How `instructions` combine with the assistant-level instructions. `replace` (default): the node's instructions are used alone. `append`: the node's instructions are concatenated after the assistant's instructions. enum: - replace - append title: Instructions Mode type: string llm_api_key_ref: description: >- Override for `Assistant.llm_api_key_ref` while this node is active. Part of the LLM bundle — see `model` for cascade semantics. title: Llm Api Key Ref type: string model: description: >- Override for `Assistant.model` while this node is active. Part of the LLM bundle (`model` + `llm_api_key_ref` + `external_llm`): when any of the three is set on the node, all three are taken from the node and the assistant-level LLM identity is not consulted. When none of the three is set, the assistant's bundle cascades unchanged. title: Model type: string name: description: Optional human-readable label, displayed in authoring UIs. title: Name type: string position: $ref: '#/components/schemas/NodePosition' description: >- Optional canvas coordinates used by authoring UIs to lay out the graph. Ignored by the runtime; round-trips so frontends can persist graph layout across reloads. shared_tool_ids: description: >- IDs of shared (org-level) tools available at this node. Knowledge bases are attached the same way — via a shared retrieval tool. Tools not listed here are not callable while this node is active. items: type: string title: Shared Tool Ids type: array tools: description: >- Full tool definitions for this node, resolved from `shared_tool_ids` server-side. Populated on responses so clients can render the flow without a follow-up fetch per shared tool. Ignored on input — set `shared_tool_ids` to configure a node's tools. items: $ref: '#/components/schemas/AssistantTools' title: Tools type: array tools_mode: default: replace description: >- How `shared_tool_ids` combine with the assistant-level tool set. `replace` (default): only the node's tools are callable. `append`: the node's tools are added to the assistant's tools. Ignored when `shared_tool_ids` is null. enum: - replace - append title: Tools Mode type: string transcription: $ref: '#/components/schemas/TranscriptionSettings' description: Per-node transcription override (response form). type: const: prompt default: prompt description: Node kind discriminator. `prompt` is an LLM-driven step. title: Type type: string voice_settings: $ref: '#/components/schemas/VoiceSettings' description: Per-node voice override (response form). required: - id - instructions title: FlowNode type: object SpeakNode: properties: type: type: string const: speak title: Type description: Node kind discriminator. Always `speak` for a speak node. default: speak id: description: Caller-supplied unique identifier for this node within the flow. title: Id type: string name: description: Optional human-readable label, displayed in authoring UIs. title: Name type: string message: type: string title: Message description: >- Message delivered to the user verbatim when the flow reaches this node. No LLM turn — the text is spoken/sent exactly as written. `{{variable}}` placeholders are interpolated from the conversation's dynamic variables; an unresolved placeholder renders as an empty string. After delivering, the flow routes via the node's outgoing `llm` / `expression` edges (commonly a single unconditional edge). position: $ref: '#/components/schemas/NodePosition' description: >- Optional canvas coordinates used by authoring UIs to lay out the graph. Ignored by the runtime; round-trips so frontends can persist graph layout across reloads. type: object required: - id - message title: SpeakNode description: A standalone scripted-message step in a flow, as returned by the API. ToolNode: description: A standalone tool step in a conversation flow, as returned by the API. properties: id: description: Caller-supplied unique identifier for this node within the flow. title: Id type: string name: description: Optional human-readable label, displayed in authoring UIs. title: Name type: string position: $ref: '#/components/schemas/NodePosition' description: >- Optional canvas coordinates used by authoring UIs to lay out the graph. Ignored by the runtime; round-trips so frontends can persist graph layout across reloads. shared_tool_id: description: >- ID of the single shared (org-level) tool this node executes. When the flow reaches this node the tool runs as a deliberate step (no LLM turn); its outgoing `tool_result` edges then route on the outcome. Arguments are filled from the conversation's dynamic variables by name — a dynamic variable whose name matches one of the tool's parameters supplies that argument. Cross-validated against the org's shared tools on write. title: Shared Tool Id type: string tool: $ref: '#/components/schemas/AssistantTools' description: >- Full tool definition resolved from `shared_tool_id` server-side. Populated on responses so clients can render the node without a follow-up fetch. Ignored on input — set `shared_tool_id`. type: const: tool default: tool description: Node kind discriminator. Always `tool` for a tool node. title: Type type: string required: - id - shared_tool_id title: ToolNode type: object WebhookToolParams: properties: name: type: string description: The name of the tool. description: type: string description: The description of the tool. url: description: >- The URL of the external tool to be called. This URL is going to be used by the assistant. The URL can be templated like: `https://example.com/api/v1/{id}`, where `{id}` is a placeholder for a value that will be provided by the assistant if `path_parameters` are provided with the `id` attribute. type: string example: https://example.com/api/v1/function method: description: The HTTP method to be used when calling the external tool. type: string enum: - GET - POST - PUT - DELETE - PATCH default: POST headers: description: The headers to be sent to the external tool. type: array items: type: object properties: name: type: string value: description: >- The value of the header. Note that we support mustache templating for the value. For example you can use `Bearer {{#integration_secret}}test-secret{{/integration_secret}}` to pass the value of the integration secret as the bearer token. [Telnyx signature headers](https://developers.telnyx.com/docs/voice/programmable-voice/voice-api-webhooks) will be automatically added to the request. type: string body_parameters: description: >- The body parameters the webhook tool accepts, described as a JSON Schema object. These parameters will be passed to the webhook as the body of the request. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema) for documentation about the format type: object properties: properties: description: The properties of the body parameters. type: object additionalProperties: true required: description: The required properties of the body parameters. type: array items: type: string type: type: string enum: - object example: properties: age: description: The age of the customer. type: integer location: description: The location of the customer. type: string required: - age - location type: object path_parameters: description: >- The path parameters the webhook tool accepts, described as a JSON Schema object. These parameters will be passed to the webhook as the path of the request if the URL contains a placeholder for a value. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema) for documentation about the format type: object properties: properties: description: The properties of the path parameters. type: object additionalProperties: true required: description: The required properties of the path parameters. type: array items: type: string type: type: string enum: - object example: properties: id: description: The id of the customer. type: string required: - id type: object query_parameters: description: >- The query parameters the webhook tool accepts, described as a JSON Schema object. These parameters will be passed to the webhook as the query of the request. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema) for documentation about the format type: object properties: properties: description: The properties of the query parameters. type: object additionalProperties: true required: description: The required properties of the query parameters. type: array items: type: string type: type: string enum: - object example: properties: page: description: The page number. type: integer required: - page type: object async: type: boolean default: false description: >- If async, the assistant will move forward without waiting for your server to respond. async_timeout_ms: type: integer minimum: 1 maximum: 15000 description: >- Maximum time in milliseconds that the conversation worker waits for an async webhook response before returning "Submitted" to the LLM. If unset, the platform default (currently 300ms) is used. timeout_ms: type: integer minimum: 500 maximum: 10000 description: >- The maximum number of milliseconds to wait for the webhook to respond. Only applicable when async is false. store_fields_as_variables: type: array description: >- A list of mappings that extract values from the webhook response and store them as dynamic variables. Each mapping specifies a dynamic variable name and a dot-notation path to the value in the response body. items: type: object required: - name - value_path properties: name: type: string minLength: 1 description: >- The name of the dynamic variable to store the extracted value in. value_path: type: string minLength: 1 description: >- A dot-notation path to the value in the webhook response body (e.g. 'customer.name' or 'id'). additionalProperties: false type: object required: - url - name - description title: WebhookToolParams BucketIds: properties: bucket_ids: items: type: string type: array description: >- List of [embedded storage buckets](https://developers.telnyx.com/api-reference/embeddings/embed-documents) to use for retrieval-augmented generation. max_num_results: description: >- The maximum number of results to retrieve as context for the language model. type: integer type: object required: - bucket_ids title: BucketIds HandoffToolParams: properties: voice_mode: type: string enum: - unified - distinct description: >- With the unified voice mode all assistants share the same voice, making the handoff transparent to the user. With the distinct voice mode all assistants retain their voice configuration, providing the experience of a conference call with a team of assistants. ai_assistants: type: array description: List of possible assistants that can receive a handoff. items: type: object properties: name: type: string description: >- Helpful name for giving context on when to handoff to the assistant. example: Scheduling Specialist id: type: string description: The ID of the assistant to hand off to. example: assistant-1234567890abcdef required: - name - id required: - ai_assistants type: object title: HandoffToolParams HangupToolParams: properties: description: type: string default: This tool is used to hang up the call. description: >- The description of the function that will be passed to the assistant. type: object title: HangupToolParams InferenceEmbeddingTransferToolParams: properties: targets: oneOf: - type: array items: type: object properties: name: type: string description: The name of the target. example: Support to: type: string description: The destination number or SIP URI of the call. example: '+13129457420' required: - to - type: string description: >- A dynamic variable string like `{{ targets }}` where `targets` is returned by the dynamic variables webhook and resolves to an array of target objects at runtime. example: '{{ targets }}' description: >- The different possible targets of the transfer. The assistant will be able to choose one of the targets to transfer the call to. This can also be a dynamic variable string like `{{ targets }}` where `targets` is returned by the dynamic variables webhook and resolves to an array of target objects at runtime. from: type: string example: '+35319605860' description: Number or SIP URI placing the call. warm_transfer_instructions: type: string example: >- Briefly greet the transfer recipient and provide any relevant information from the call. Let them know you will bridge the call right after. description: >- Natural language instructions for your agent for how to provide context for the transfer recipient. description: type: string description: >- A description of the transfer tool. By default, Telnyx generates this automatically based on the configured targets. Typically only set when importing an assistant from another provider that allowed a custom description; in that case the provided value is preserved. Most users should leave this empty and let Telnyx manage it. example: Transfer the call to a human agent. warm_message_delay_ms: type: - integer - 'null' example: 2000 description: >- Optional delay in milliseconds before playing the warm message audio when the transferred call is answered. When set, the audio_url is not included in the dial command; instead, playback starts after the specified delay. When not set, existing behavior (audio_url in dial) is preserved. custom_headers: description: >- Custom headers to be added to the SIP INVITE for the transfer command. type: array items: type: object properties: name: type: string value: description: >- The value of the header. Note that we support mustache templating for the value. For example you can use `{{#integration_secret}}test-secret{{/integration_secret}}` to pass the value of the integration secret. type: string voicemail_detection: type: object description: >- Configuration for voicemail detection (AMD - Answering Machine Detection) on the transferred call. Allows the assistant to detect when a voicemail system answers the transferred call and take appropriate action. properties: detection_mode: type: string enum: - disabled - premium description: >- The AMD detection mode to use. 'premium' enables premium answering machine detection. 'disabled' turns off AMD detection. on_voicemail_detected: type: object description: >- Action to take when voicemail is detected on the transferred call. properties: action: type: string enum: - stop_transfer - leave_message_and_stop_transfer description: >- The action to take when voicemail is detected. 'stop_transfer' hangs up immediately. 'leave_message_and_stop_transfer' leaves a message then hangs up. voicemail_message: type: object description: >- Configuration for the voicemail message to leave. Only applicable when action is 'leave_message_and_stop_transfer'. properties: type: type: string enum: - message - warm_transfer_instructions description: >- The type of voicemail message. Use 'message' to leave a specific TTS message, or 'warm_transfer_instructions' to play the warm transfer audio. message: type: string description: >- The specific message to leave as voicemail (converted to speech). Only applicable when type is 'message'. detection_config: type: object description: >- Advanced AMD detection configuration parameters. All values are optional - Telnyx will use defaults if not specified. properties: greeting_silence_duration_millis: type: integer minimum: 500 maximum: 50000 description: >- Duration of silence after the greeting to wait before considering the greeting complete. greeting_total_analysis_time_millis: type: integer minimum: 500 maximum: 300000 description: Maximum time to spend analyzing the greeting. after_greeting_silence_millis: type: integer minimum: 100 maximum: 5000 description: >- Duration of silence after greeting detection before finalizing the result. between_words_silence_millis: type: integer minimum: 10 maximum: 5000 description: Maximum silence duration between words during greeting. greeting_duration_millis: type: integer minimum: 100 maximum: 10000 description: Expected duration of greeting speech. initial_silence_millis: type: integer minimum: 100 maximum: 500000 description: >- Maximum silence duration at the start of the call before speech. maximum_number_of_words: type: integer minimum: 1 maximum: 50 description: Maximum number of words expected in a human greeting. maximum_word_length_millis: type: integer minimum: 50 maximum: 5000 description: Maximum duration of a single word. silence_threshold: type: integer minimum: 1 maximum: 1000 description: Audio level threshold for silence detection. total_analysis_time_millis: type: integer minimum: 500 maximum: 60000 description: Total time allowed for AMD analysis. min_word_length_millis: type: integer minimum: 1 maximum: 1000 description: Minimum duration for audio to be considered a word. type: object required: - targets - from title: TransferToolParams InviteToolConfig: properties: from: type: string example: '+35319605860' description: Number or SIP URI placing the call. targets: oneOf: - type: array items: type: object properties: name: type: string description: The name of the target. example: Support to: type: string description: The destination number or SIP URI of the call. example: '+13129457420' required: - to - type: string description: >- A dynamic variable string like `{{ targets }}` where `targets` is returned by the dynamic variables webhook and resolves to an array of target objects at runtime. example: '{{ targets }}' - type: 'null' description: >- The different possible targets of the invite. The assistant will be able to choose one of the targets to invite to the call. This can also be a dynamic variable string like `{{ targets }}` where `targets` is returned by the dynamic variables webhook and resolves to an array of target objects at runtime. If omitted or null, the invite tool can still be configured and targets may be supplied dynamically at runtime. custom_headers: description: Custom headers to be added to the SIP INVITE for the invite command. type: array items: type: object properties: name: type: string value: description: >- The value of the header. Note that we support mustache templating for the value. For example you can use `{{#integration_secret}}test-secret{{/integration_secret}}` to pass the value of the integration secret. type: string voicemail_detection: type: object description: >- Configuration for voicemail detection (AMD - Answering Machine Detection) on the invited call. properties: detection_mode: type: string enum: - disabled - premium description: >- The AMD detection mode to use. 'premium' enables premium answering machine detection. 'disabled' turns off AMD detection. on_voicemail_detected: type: object description: Action to take when voicemail is detected on the invited call. properties: action: type: string enum: - stop_invite description: The action to take when voicemail is detected. type: object required: - from title: InviteToolConfig SIPReferToolParams: properties: targets: type: array description: >- The different possible targets of the SIP refer. The assistant will be able to choose one of the targets to refer the call to. items: type: object properties: name: type: string description: The name of the target. example: Support sip_address: type: string description: The SIP URI to which the call will be referred. example: sip:username@sip.non-telnyx-address.com sip_auth_username: type: string description: SIP Authentication username used for SIP challenges. sip_auth_password: type: string description: SIP Authentication password used for SIP challenges. required: - name - sip_address sip_headers: description: >- SIP headers to be added to the SIP REFER. Currently only User-to-User and Diversion headers are supported. type: array items: type: object properties: name: type: string enum: - User-to-User - Diversion value: description: >- The value of the header. Note that we support mustache templating for the value. For example you can use `{{#integration_secret}}test-secret{{/integration_secret}}` to pass the value of the integration secret. type: string custom_headers: description: Custom headers to be added to the SIP REFER. type: array items: type: object properties: name: type: string value: description: >- The value of the header. Note that we support mustache templating for the value. For example you can use `{{#integration_secret}}test-secret{{/integration_secret}}` to pass the value of the integration secret. type: string type: object required: - targets title: SIPReferToolParams SkipTurnToolParams: properties: description: type: string default: >- This tool is used to skip the assistant turn without producing a response. description: >- The description of the function that will be passed to the assistant. type: object title: SkipTurnToolParams TranscriptionEndpointingPlan: type: object title: TranscriptionEndpointingPlan description: >- Endpointing thresholds used to decide when the user has finished speaking. Applies to non turn-taking transcription models. For `deepgram/flux`, use `transcription.settings.eot_threshold` / `eot_timeout_ms` / `eager_eot_threshold`. properties: on_punctuation_seconds: type: number format: float default: 0.1 description: Seconds to wait after the transcript ends with punctuation. on_no_punctuation_seconds: type: number format: float default: 1.5 description: Seconds to wait after the transcript ends without punctuation. on_number_seconds: type: number format: float default: 0.5 description: Seconds to wait after the transcript ends with a number. ExpressionCondition: description: |- Edge condition evaluated as a deterministic expression AST. The expression is computed against runtime dynamic variables and must evaluate to a boolean. Prefer this over `LLMCondition` when the rule is a clean function of known variables — it's cheaper and predictable. example: expression: left: name: user_age type: variable op: '>=' right: type: number_literal value: 18 type: comparison type: expression properties: expression: $ref: '#/components/schemas/Expression' type: const: expression title: Type type: string required: - type - expression title: ExpressionCondition type: object LLMCondition: description: >- Edge condition evaluated by the LLM from a natural-language prompt. The model is asked to judge the prompt against conversation context and returns true/false. Use this for fuzzy intents that aren't expressible as a deterministic expression (e.g. 'user wants to escalate to a human'). example: prompt: The user has confirmed their booking details. type: llm properties: prompt: description: Natural-language criterion the LLM judges as true/false. example: The user has confirmed their booking details. title: Prompt type: string type: const: llm title: Type type: string required: - type - prompt title: LLMCondition type: object DefaultCondition: properties: type: type: string const: default title: Type type: object required: - type title: DefaultCondition description: >- Fallback edge condition: fires only when no other edge's condition is true. Evaluated after every conditioned (`llm` / `expression`) edge regardless of declaration order, so it routes the flow whenever none of the node's other outgoing edges match. Valid **only** on edges leaving a `tool` or `speak` node, where the deterministic step auto-advances and must always have somewhere to go. A tool/speak node with any outgoing edge is required to carry exactly one `default` edge so it never dead-ends; a tool/speak node with no outgoing edges is a valid terminal step. Carries no parameters. example: type: default AssistantTarget: description: |- Edge target referencing a different assistant. When the edge fires, the conversation hands off to `assistant_id`: the active assistant on the conversation row is rewritten and the new assistant's flow starts at its own `start_node_id`. The current turn's LLM response is delivered to the user as-is; subsequent turns route to the new assistant. properties: assistant_id: description: ID of the assistant the conversation transitions to. example: assistant-billing title: Assistant Id type: string position: $ref: '#/components/schemas/NodePosition' description: >- Optional canvas coordinates for rendering the target assistant as a node in authoring UIs. Pure presentation — the runtime ignores it; round-trips so frontends can persist graph layout across reloads. When multiple edges target the same assistant, each edge's `position` is independent (frontends typically use the first non-null one). example: x: 600 'y': 80 type: const: assistant title: Type type: string voice_mode: default: unified description: >- Voice behavior when handing off to the target assistant, mirroring the handoff tool's `voice_mode`. `unified` (default) keeps the current voice across the handoff; `distinct` lets the target assistant speak with its own configured voice. Only applies to assistant targets — node targets override voice via the node's own `voice_settings`. enum: - unified - distinct title: Voice Mode type: string required: - type - assistant_id title: AssistantTarget type: object NodeTarget: description: |- Edge target referencing another node within the same flow. The runtime transitions the active node to `node_id` and continues processing within the current assistant's flow. properties: node_id: description: ID of the node this edge transitions into. example: n_billing title: Node Id type: string type: const: node title: Type type: string required: - type - node_id title: NodeTarget type: object NodePosition: description: >- 2D coordinates for a node, used by authoring UIs to lay out the graph. Purely a presentation aid. The runtime ignores `position`; it round-trips through the API so frontends can persist the graph layout customers arrange in the editor. properties: x: description: Horizontal coordinate in the authoring canvas. title: X type: number 'y': description: Vertical coordinate in the authoring canvas. title: 'Y' type: number required: - x - 'y' title: NodePosition type: object Expression: title: Expression description: >- A node in a deterministic expression AST. Exactly one variant is selected by the `type` discriminator. Terminal variants (`number_literal`, `string_literal`, `bool_literal`, `variable`) bottom out the recursion; `arithmetic`, `bool_op`, and `comparison` nest further sub-expressions. Extracted into a single named schema so the recursive union is defined once (was previously inlined at every operand site). oneOf: - $ref: '#/components/schemas/ComparisonExpression' - $ref: '#/components/schemas/BooleanOpExpression' - $ref: '#/components/schemas/ArithmeticExpression' - $ref: '#/components/schemas/DynamicVariableExpression' - $ref: '#/components/schemas/StringLiteralExpression' - $ref: '#/components/schemas/NumberLiteralExpression' - $ref: '#/components/schemas/BooleanLiteralExpression' discriminator: mapping: arithmetic: $ref: '#/components/schemas/ArithmeticExpression' bool_literal: $ref: '#/components/schemas/BooleanLiteralExpression' bool_op: $ref: '#/components/schemas/BooleanOpExpression' comparison: $ref: '#/components/schemas/ComparisonExpression' number_literal: $ref: '#/components/schemas/NumberLiteralExpression' string_literal: $ref: '#/components/schemas/StringLiteralExpression' variable: $ref: '#/components/schemas/DynamicVariableExpression' propertyName: type ComparisonExpression: description: |- Compare two sub-expressions with a relational or membership operator. Evaluates to a boolean. Used in edge conditions to gate transitions on runtime values, e.g. `user_age >= 18` or `tier == "gold"`. example: left: name: user_age type: variable op: '>=' right: type: number_literal value: 18 type: comparison properties: left: $ref: '#/components/schemas/Expression' op: description: >- Relational/membership operator. `contains` / `not_contains` apply to strings (substring) and arrays (membership). enum: - '==' - '!=' - < - <= - '>' - '>=' - contains - not_contains title: Op type: string right: $ref: '#/components/schemas/Expression' type: const: comparison title: Type type: string required: - type - op - left - right title: ComparisonExpression type: object BooleanOpExpression: description: |- Combine sub-expressions with a logical operator (`and` / `or` / `not`). `and` and `or` accept two or more operands; `not` accepts exactly one. example: op: and operands: - name: is_premium type: variable - left: name: user_age type: variable op: '>=' right: type: number_literal value: 18 type: comparison type: bool_op properties: op: description: Logical operator. `not` is unary; `and`/`or` are n-ary (>=2). enum: - and - or - not title: Op type: string operands: description: >- Operand sub-expressions. Length must be exactly 1 for `not` and >= 2 for `and`/`or`. items: $ref: '#/components/schemas/Expression' title: Operands type: array type: const: bool_op title: Type type: string required: - type - op - operands title: BooleanOpExpression type: object ArithmeticExpression: description: >- Numeric expression: applies an arithmetic operator to two sub-expressions. Useful for derived numeric checks, e.g. `cart_total + shipping > 50`. Both operands should resolve to numbers at runtime. example: left: name: cart_total type: variable op: + right: name: shipping_cost type: variable type: arithmetic properties: left: $ref: '#/components/schemas/Expression' op: description: Arithmetic operator applied to `left` and `right`. enum: - + - '-' - '*' - / - '%' title: Op type: string right: $ref: '#/components/schemas/Expression' type: const: arithmetic title: Type type: string required: - type - op - left - right title: ArithmeticExpression type: object DynamicVariableExpression: description: |- Reference a dynamic variable by name. Resolved at runtime from the assistant's dynamic-variables context (see `Assistant.dynamic_variables` and the dynamic-variables webhook). example: name: user_age type: variable properties: name: description: Variable name to look up in the runtime context. example: user_age title: Name type: string type: const: variable title: Type type: string required: - type - name title: DynamicVariableExpression type: object StringLiteralExpression: description: Constant string value. example: type: string_literal value: gold properties: type: const: string_literal title: Type type: string value: description: Literal string value. example: gold title: Value type: string required: - type - value title: StringLiteralExpression type: object NumberLiteralExpression: description: >- Constant numeric value (float; integers are accepted and stored as float). example: type: number_literal value: 18 properties: type: const: number_literal title: Type type: string value: description: Literal numeric value. example: 18 title: Value type: number required: - type - value title: NumberLiteralExpression type: object BooleanLiteralExpression: description: Constant boolean value. Useful for unconditional ('always') edges. example: type: bool_literal value: true properties: type: const: bool_literal title: Type type: string value: description: Literal boolean value. title: Value type: boolean required: - type - value title: BooleanLiteralExpression type: object securitySchemes: bearerAuth: scheme: bearer type: http ````