> ## Documentation Index > Fetch the complete documentation index at: https://docs.convocore.ai/llms.txt > Use this file to discover all available pages before exploring further. # Send Outbound SMS > Send an SMS from an agent's connected Twilio number — direct message or AI-generated opener. ## Example — direct send ```json theme={null} { "agentId": "your-agent-id", "to": "+15551234567", "message": "Hi Jane — following up on your appointment.", "campaignId": "campaign-123", "leadInfo": { "username": "Jane", "email": "jane@example.com" } } ``` ## Example — AI-generated opener (`mode: agent`) ```json theme={null} { "agentId": "your-agent-id", "to": "+15551234567", "mode": "agent", "leadInfo": { "username": "Jane", "company": "Acme Corp" }, "options": { "initialPrompt": "Write a short friendly SMS to remind them about tomorrow's appointment.", "sendViaEdge": true } } ``` `to` and optional `from` must be E.164 (`+` and country code). The agent must have an SMS-capable Twilio number connected (own account or platform rental). Use `options.dryRun: true` to validate routing (from number, mode, segment count) without sending. ## OpenAPI ````yaml POST /sms openapi: 3.0.3 info: title: Convocore OpenAPI description: Full API reference for Convocore version: 1.0.4 servers: - url: https://eu-gcp-api.vg-stuff.com/v3 description: EU Node.js API server - url: https://na-gcp-api.vg-stuff.com/v3 description: NA Node.js API server security: [] paths: /sms: post: tags: - Numbers and Telephony summary: Send an outbound SMS from an agent's connected number description: >- Initiate an outbound SMS using a Twilio number already connected to the agent. Supports direct sends, AI-generated opener messages, MMS media, campaign/lead metadata, and advanced Twilio delivery options. operationId: sms parameters: [] requestBody: required: true content: application/json: schema: type: object properties: agentId: type: string minLength: 1 description: Agent ID whose connected SMS number will send the message. to: type: string pattern: ^\+ description: Recipient phone number in E.164 format, e.g. +15551234567. from: type: string pattern: ^\+ description: >- Specific connected Twilio number to send from. Defaults to the agent's primary SMS number. message: type: string description: >- Message body for direct mode. Required unless mode is agent and initialPrompt is provided via options. mediaUrl: type: string format: uri description: Single MMS media URL. mediaUrls: type: array items: type: string format: uri maxItems: 10 description: Multiple MMS media URLs (Twilio supports up to 10). mode: type: string enum: - direct - agent default: direct description: >- direct = send the message as-is. agent = run the agent first and send its generated reply. campaignId: type: string description: >- Optional campaign ID to attach to the outbound SMS conversation. convoId: type: string description: >- Custom conversation ID. Defaults to the normalized recipient phone number digits. leadInfo: type: object properties: username: type: string email: type: string address: type: string additionalProperties: true description: >- Lead metadata merged into the conversation for personalization. options: type: object properties: dryRun: type: boolean description: >- Validate and return the resolved send plan without sending. createConvo: type: boolean description: >- Create or update the SMS conversation record. Defaults to true. persistToConvo: type: boolean description: >- Persist outbound message metadata on the conversation. Defaults to true. skipIfHumanHandoff: type: boolean description: >- Skip sending when the conversation is in human-chatting handoff state. splitLongMessages: type: boolean description: >- Split messages longer than 1600 chars into multiple SMS segments. sendViaEdge: type: boolean description: >- Deliver through the edge SMS sender so transcripts and multi-segment replies stay consistent. disableRecordHistory: type: boolean description: >- When using agent mode, skip saving the agent turn to Firebase history. initialPrompt: type: string description: >- Prompt passed to the agent in agent mode. Defaults to a generic outbound SMS opener. statusCallbackUrl: type: string format: uri description: Override Twilio delivery status callback URL. messagingServiceSid: type: string description: Twilio Messaging Service SID (own-account mode only). validityPeriod: type: integer minimum: 1 maximum: 14400 description: >- Twilio validity period in seconds (own-account mode only). provideFeedback: type: boolean description: >- Request delivery feedback from Twilio (own-account mode only). smartEncoded: type: boolean description: >- Enable Twilio smart encoding for GSM-7/Unicode optimization. agentOptions: type: object properties: serverData: type: object properties: agentTools: type: array agentVariables: type: array agentOrg: {} agentClients: type: array orgTeams: type: array agentData: type: object properties: title: type: string description: '**Title**: The title of the agent.' description: type: string description: >- **Description**: A brief description of the agent. ownerID: type: string description: >- **Owner ID**: The unique identifier of the agent's owner. theme: type: string description: >- **Theme**: The visual theme of the agent, e.g., 'blue-light' or 'custom-blue-dark'. voiceConfig: type: object properties: transcriber: type: object properties: speechConfig: type: object properties: format: type: string sampleRate: type: number language: type: string additionalProperties: false modelId: type: string patienceFactor: type: number language: type: string provider: type: string enum: - deepgram - gladia - assemblyai - speechmatics - google-cloud-speech - google-live-transcription randomOptions: {} internal: type: object properties: inputAudioStream: {} debug: type: boolean additionalProperties: false apiKey: type: string platformSpecific: type: object properties: deepgram: type: object properties: keywords: type: array items: type: string description: >- Keywords to focus on in the transcription. language: type: string smart_format: type: boolean model: type: string interim_results: type: boolean endpointing: type: number no_delay: type: boolean autoLanguageDetection: type: boolean additionalProperties: false assemblyai: type: object properties: modelId: type: string required: - modelId additionalProperties: false googleCloud: type: object properties: keywords: type: array items: type: string description: >- Keywords or phrases to focus on in the transcription. model: type: string description: The Google Cloud Speech model to use. languageCode: type: string description: The language code for transcription. enableAutomaticPunctuation: type: boolean description: Enable automatic punctuation. enableWordTimeOffsets: type: boolean description: Enable word time offsets. maxAlternatives: type: number description: >- Maximum number of recognition alternatives. profanityFilter: type: boolean description: Enable profanity filter. speechContexts: type: array items: type: object properties: phrases: type: array items: type: string boost: type: number required: - phrases additionalProperties: false description: Speech contexts for better recognition. additionalProperties: false additionalProperties: false utteranceThreshold: type: number inputVoiceEnhancer: type: boolean required: - provider additionalProperties: false description: >- **Transcriber**: The configuration options for the transcriber provider used by the agent. speechGen: allOf: - type: object properties: highAudioQuality: type: boolean description: >- Generate highest quality audio possible, must have pro plan or higher on elevenlabs to enable your own key to work with this option. provider: type: string enum: - elevenlabs - deepgram - cartesia - rime-ai - openai - google-cloud - google-live - ultravox - grok-live - minimax description: Speech providers supported by the app. modelId: type: string description: The model id of the speech gen. voiceId: type: string description: The voice id for the speech service. apiKey: type: string region: type: string backgroundNoise: type: string enum: - restaurant - office - park - street punctuationBreaks: type: array items: type: string platformSpecific: type: object properties: elevenLabs: type: object properties: stability: type: number similarity_boost: type: number use_speaker_boost: type: boolean speed: type: number style: type: number additionalProperties: false playht: type: object properties: voiceId: type: string required: - voiceId additionalProperties: false cartesia: type: object properties: emotion: type: string enum: - neutral - calm - angry - content - sad - scared speed: type: number volume: type: number promptInstructions: type: string additionalProperties: false ultravox: type: object properties: temperature: type: number minimum: 0 maximum: 1 speed: type: number minimum: 0.7 maximum: 1.2 voiceOverridesProvider: type: string enum: - elevenLabs - cartesia - google - inworld - lmnt elevenLabsModel: type: string vadSettings: type: object properties: turnEndpointDelay: type: string minimumTurnDuration: type: string minimumInterruptionDuration: type: string frameActivationThreshold: type: number minimum: 0.1 maximum: 1 additionalProperties: false additionalProperties: false grokLive: type: object properties: turnDetection: type: object properties: type: type: string enum: - server_vad threshold: type: number minimum: 0.1 maximum: 0.9 silence_duration_ms: type: number minimum: 0 prefix_padding_ms: type: number minimum: 0 additionalProperties: false inputAudioTranscription: type: object properties: model: type: string default: grok-2-audio additionalProperties: false audio: type: object properties: input: type: object properties: format: type: object properties: type: type: string enum: - audio/pcm - audio/pcmu - audio/pcma rate: type: number required: - type additionalProperties: false additionalProperties: false output: type: object properties: format: type: object properties: type: type: string enum: - audio/pcm - audio/pcmu - audio/pcma rate: type: number required: - type additionalProperties: false additionalProperties: false additionalProperties: false enableWebSearch: type: boolean enableXSearch: type: boolean debug: type: boolean additionalProperties: false additionalProperties: false backChannelling: type: boolean description: >- Whether to say umm, sure, etc.. when the user is talking for too long. language: type: string enableLongMessageBackchannelling: type: boolean description: >- Whether to say umm, sure, etc.. when the user is talking for too long. backchannelMessages: type: array items: type: string backchannelInterval: type: number wordsReplacements: type: array items: type: object properties: word: type: string replacement: type: string required: - word - replacement additionalProperties: false required: - provider - type: object properties: internal: type: object properties: inputTextStream: {} audioConfig: type: object properties: format: type: string sampleRate: type: number channels: type: number language: type: string required: - format - sampleRate - channels additionalProperties: false emitOnComplete: type: boolean additionalProperties: false description: >- **Speech Generation**: The configuration options for the speech generation provider used by the agent. config: type: object properties: recordAudio: type: boolean backgroundNoise: type: string enum: - restaurant - office - street - none enableWebCalling: type: boolean firstInputChunkUNIXMs: type: number firstOutputChunkUNIXMs: type: number required: - recordAudio additionalProperties: false description: >- **Call Configuration**: The call configuration settings for the agent. additionalProperties: false light: type: boolean description: >- Whether to enable light mode for the agent. If set to true, the agent will not retain previous chat interactions for privacy reasons. enableVertex: type: boolean description: Whether to enable Vertex AI for the agent. autoOpenWidget: type: boolean description: >- Whether to auto-open the widget when the agent is loaded. createdAtUNIX: type: number description: >- The timestamp when the agent was created, formatted as an ISO 8601 string. Useful for tracking tool lifecycle. disabled: type: boolean description: >- Indicates whether the agent is disabled. If set to true, the agent will not be active. vectorDb: description: >- Specifies the vector database configuration for the agent. This database is used for handling vectorized data or embeddings. withRefresh: type: boolean description: >- Determines whether the agent should refresh its settings or data dynamically. agentPlatform: type: string description: >- Specifies the platform on which the agent operates, such as web, mobile, or other environments. ID: type: string description: >- A unique identifier for the agent used internally. storageID: type: string description: >- The identifier for the agent's storage configuration. VF_DIALOGUE_API_KEY: type: string description: >- API key for VF dialogue integration, enabling the agent to communicate with external systems. VF_PROJECT_API_KEY: type: string description: >- API key associated with the VF project to authenticate and enable specific functionalities. VF_PROJECT_ID: type: string description: >- The unique identifier of the VF project linked to the agent. VF_ENVIRONMENT_ALIAS: type: string description: >- Voiceflow environment alias used for v4 session bootstrap, typically `main`. VF_KB_API_KEY: type: string description: >- API key for accessing the VF knowledge base, enabling advanced knowledge retrieval capabilities. voiceflowRuntimeMode: type: string enum: - legacy - auto - v4 description: >- Controls which Voiceflow runtime path the agent should use. roundedImageURL: type: string description: >- URL pointing to the rounded image asset used to visually represent the agent. rectangeImageURL: type: string description: >- URL pointing to the rectangular image asset for the agent's branding or display. messageDelayMS: type: number description: >- Specifies the delay in milliseconds for the agent to send messages, simulating natural conversation timing. scrollAnimation: type: boolean description: >- Indicates whether scroll animations are enabled for smooth transitions in the chat interface. proactiveMessage: type: string description: >- A predefined message the agent sends proactively to engage users. acceptFileUpload: type: boolean description: >- Determines if the agent accepts file uploads from users as part of the interaction. recordChatHistory: type: boolean description: >- Specifies whether the agent logs chat histories for future reference or analysis. chatBgURL: type: string description: >- URL of the background image used in the chat interface. disableSmoothScroll: type: boolean description: >- Disables smooth scrolling if set to true, potentially improving performance on low-end devices. isDeployed: type: boolean description: >- Indicates whether the agent is currently deployed and operational. tokensUsage: description: >- Tracks the token usage statistics for the agent, typically used in LLM integrations. maxTokensUsage: description: >- Defines the maximum allowable token usage for the agent to prevent overconsumption. lastModified: type: number description: >- The timestamp of the last modification made to the agent's configuration. fontFamily: type: string description: >- The font family used in the agent's user interface for consistent branding. branding: type: string description: >- Custom branding information or labels associated with the agent. customThemeJSONString: type: string description: >- A JSON string defining a custom theme for the agent's interface. autoStartWidget: type: boolean description: >- Determines if the widget should auto-start upon page load. allTimeTriggers: type: number description: >- The total number of triggers or interactions the agent has handled. syncBrowser: type: boolean description: >- Enables synchronization of settings or states with the browser session. delayBeforeSubmit: type: number description: >- The delay (in milliseconds) before submitting user input to the agent. messageAggregationSeconds: type: number description: >- Number of seconds the AI should wait and aggregate consecutive user messages before triggering a single reply. Applies globally to all text-based channels (Web, WhatsApp, Instagram, Messenger, Telegram, SMS, Discord). Default 0 = reply instantly after every message (current behaviour). When > 0, every new inbound message resets the wait timer (debounced); after the window of silence elapses, all buffered messages are concatenated into a single user turn and the AI is invoked once. region: type: string enum: - eu - na description: >- Defines the geographical region settings for the agent, such as localization. listenForUrlChanges: type: boolean description: >- Specifies if the agent should monitor and react to changes in the browser URL. chatForget: type: boolean description: >- If enabled, the agent will not retain previous chat interactions for privacy reasons. lang: type: string description: >- Specifies the default language for the agent's interactions and responses. enableAudioSupport: type: boolean description: >- Enables audio features, allowing the agent to process and respond to audio inputs. AITranslateTo: type: string description: >- Defines the target language for AI translation during interactions. enableAITranslate: type: boolean description: >- Activates AI-based translation for multilingual support. disableNoReplyListener: type: boolean description: >- Disables the listener for instances where no reply is detected from the agent. disableHumanHandoff: type: boolean description: >- Prevents the agent from escalating interactions to a human operator. alwaysShowHandoff: type: boolean description: >- Ensures the handoff option is always visible, regardless of agent status. manualControl: type: boolean description: >- Allows manual control over specific agent functionalities. enableVGHandoff: type: boolean description: >- Activates VG-specific handoff mechanisms for the agent. enableGeoAnalytics: type: boolean description: >- Enables geographic analytics to track user interactions based on location. hideVoiceflowAnalytics: type: boolean description: >- Hides Voiceflow-specific analytics including intents, conversation flow metrics, and rating analytics. hideCallAnalytics: type: boolean description: >- Hides voice call analytics including duration, latency, cost breakdown, call counts, and call end reasons. hideConversationMetrics: type: boolean description: >- Hides conversation metrics including user retention, conversation length, time retention, and engagement metrics. hideCustomAnalytics: type: boolean description: >- Hides user-created custom charts and metrics from the analytics dashboard. hideVapiAnalytics: type: boolean description: >- Hides VAPI-specific analytics including cost tracking and voice-related metrics. hideFunnelAnalytics: type: boolean description: >- Hides the user engagement funnel chart that tracks progression from widget views to high engagement. hideHandoffAnalytics: type: boolean description: >- Hides handoff analytics including accepted handover count, average response time, and average handling time metrics. fixedHandoffPopup: type: boolean description: >- Ensures the handoff popup remains fixed in the interface. buttonsLayout: type: string enum: - horizontal - vertical - in-footer description: >- Defines the layout and style of buttons in the agent's interface. ADVANCED_customCSS: type: string description: >- Specifies advanced custom CSS for overriding the default styling of the agent's interface. messagesLimit: type: number description: >- Sets a cap on the number of messages the agent can exchange in a single session. whatsappToken: type: string description: >- Token for authenticating and integrating with WhatsApp services. whatsappNumberId: type: string description: >- The ID associated with the WhatsApp number used by the agent. whatsappBusniessId: type: string description: >- The business ID for WhatsApp integration, enabling enterprise-level features. whatsappCustomBridge: type: object properties: enabled: type: boolean description: >- When enabled, inbound WhatsApp webhooks are forwarded to a developer-provided server and the returned payload is sent back through WhatsApp. webhookUrl: type: string description: >- Developer webhook URL that receives raw WhatsApp events and returns the outgoing WhatsApp payload to send. forwardToDeveloper: type: boolean description: >- When enabled, the raw Meta webhook payload is forwarded to the developer URL with zero alteration. The developer response is sent directly to the WhatsApp API as-is, with no normalization or processing. additionalProperties: false description: >- Developer-focused WhatsApp bridge that bypasses the built-in agent reply flow. waTestNumber: type: string description: >- A test number used during WhatsApp integration setup or testing. waVerifyPassed: type: boolean description: >- Indicates if WhatsApp verification was successfully completed. waTestPassed: type: boolean description: >- Indicates if the WhatsApp integration test was successful. twilioSmsAccountSid: type: string description: Twilio Account SID for SMS integration. twilioSmsAuthToken: type: string description: >- Twilio Auth Token for SMS integration (encrypted). twilioSmsPhoneNumber: type: string description: Twilio phone number for SMS integration. twilioSmsMode: type: string enum: - own_account - platform_rental description: >- Mode for Twilio SMS - using own account or platform rental. twilioSmsVerified: type: boolean description: >- Indicates if the Twilio SMS integration is verified. twilioSmsPhoneNumberSid: type: string description: >- Twilio IncomingPhoneNumber SID for platform rental numbers (used for releasing). twilioSmsRentalSubscriptionId: type: string description: >- Stripe subscription ID for platform rental SMS number billing. twilioSmsRentalStatus: type: string enum: - active - pending - cancelled - past_due description: >- Status of the SMS number rental subscription. webhookUrl: type: string description: >- The URL where the agent sends webhook events to external systems. gcloudPrivateKey: type: string description: >- Private key for Google Cloud services integration. gcloudClientEmail: type: string description: >- Client email for Google Cloud authentication. smartInit: type: boolean description: >- Activates smart initialization features for the agent. UIhandoffTitle: type: string description: >- Title displayed during UI handoff interactions. UIhandoffSubtitle: type: string description: >- Subtitle displayed during UI handoff interactions. igVerified: type: boolean description: >- Indicates if the agent's Instagram integration is verified. igAccessToken: type: string description: >- Access token for authenticating Instagram API requests. showHandoffEvenIfOffline: type: boolean description: >- Displays the handoff option even if the agent is offline. persistHandoffOnRefresh: type: boolean description: >- Maintains handoff state when users refresh or close their browser tabs. enableBugReporting: type: boolean description: >- Enables bug reporting module in the conversation's controls. blockInputDuringHandoffRequest: type: boolean description: >- Blocks chat input when a user requests human handoff but no human has been assigned yet. Only applies to VF agents. disableHandoffMessages: type: boolean description: >- When enabled, the system will not send automatic messages when handoff occurs (e.g., 'Now chatting with John'). disableHandoffStatusMessages: type: boolean description: >- When enabled, the system will not send status messages during handoff transitions (e.g., 'We are connecting you to a human', 'Handoff cancelled', 'No agent was available'). customButtonJSON_STRING: type: string description: >- JSON string defining custom buttons for the agent's interface. vg_initPrompt: type: string description: >- The initial prompt displayed or used by the VF system. vg_prompt: type: string description: >- The main prompt template used for generating responses. vg_initMessages: type: array items: type: string description: >- Initial messages loaded for the agent's conversation flow. vg_systemPrompt: type: string description: >- System-level prompt defining the agent's behavior and rules. vg_temperature: type: number description: >- Defines the temperature setting for response variability in the VF model. vg_defaultModel: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm description: >- Specifies the default LLM model used by the agent for language processing. vg_maxTokens: type: number description: >- Sets the maximum token limit for generating responses. vg_kbDefaultDimension: type: number description: >- Default dimensionality for the agent's knowledge base embeddings. vg_kbTopChunks: type: number description: >- The number of top chunks retrieved from the knowledge base during a query. SECRET_API_KEY: type: string description: >- The secret API key for secure integration with VF services. vg_kbCharCount: type: number description: >- Tracks the total character count of the agent's knowledge base. vg_kbDocsNum: type: number description: >- The total number of documents stored in the agent's knowledge base. vg_enableUIEngine: type: boolean description: >- Activates the UI engine for enhanced interface capabilities. vg_enableAboutContext: type: boolean description: >- Enables contextual information about the agent for users. vg_kb_llm: type: object properties: searchMethod: anyOf: - type: string enum: - smart - type: string enum: - simple searchPrompt: type: string modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temprature: type: number maxTokens: type: number additionalProperties: false description: >- Configuration for LLM-based knowledge base queries, including search methods and model parameters. vg_enableUIEngineForms: type: boolean description: >- Enables form and input components in the UI engine for web channel. vg_max_messages_history: type: number description: >- Maximum number of messages retained in the chat history. ifKnowsThreshold: type: number description: >- Confidence threshold for determining if the agent 'knows' an answer. avatarImageUrl: type: string description: >- URL of the avatar image representing the agent. headerImageUrl: type: string description: >- URL of the header image displayed in the chat interface. bannerImageUrl: type: string description: >- URL of the banner image used in the agent's branding. soundEffectUrl: type: string description: >- URL of the sound effect file used for agent interactions. soundEffectEnabled: type: boolean description: >- Enables or disables sound effects for the agent. enableQuickFileUpload: type: boolean description: >- Allows quick file uploads during chat interactions. instagramOptions: type: object properties: pageId: type: string accessToken: type: string additionalProperties: false translateUserResponse: type: boolean starred: type: boolean flowise: type: object properties: webhookUrl: type: string webhookSecret: type: string additionalProperties: false vfSettings: type: object properties: enableIgnoreUrlPaths: type: boolean additionalProperties: false vapi: {} vapiConfig: type: object properties: syncKbOnSave: type: boolean PUBLIC_API_KEY: type: string PRIVATE_API_KEY: type: string enableVapiOnWeb: type: boolean overrideWithVG: type: boolean promptOnWeb: type: boolean maxCostMonthly: type: number maxMinutesMonthly: type: number useVfProject: type: boolean syncTools: type: boolean additionalProperties: false vgOptions: type: object properties: isLlmStudio: type: boolean showSources: type: boolean lightUiEngine: type: boolean legacyKb: type: boolean maxChunkSize: type: number handoffTimeout: type: number additionalProperties: false ui: type: object properties: showWhatsappButtonOnWeb: type: boolean description: >- Show little whatsapp button on web if it's connected. bgImageVisible: type: boolean widgetType: anyOf: - type: string enum: - direct-chat - type: string enum: - tabs enableFixedFeedbackBtns: type: boolean switchConversationSides: type: boolean switchDashboardOnly: type: boolean voice: type: object properties: question: type: string continueBtnLabel: type: string dismissBtnLabel: type: string disableVoiceOrb: type: boolean voiceOnlyMode: type: boolean disableChatWhenVoiceOpen: type: boolean required: - question - continueBtnLabel - dismissBtnLabel additionalProperties: false additionalProperties: false assignedToolsIds: type: array items: type: string tools: type: array items: type: object properties: isDefault: type: boolean description: >- Indicates whether this tool is the default tool for the agent. Defaults to false if unspecified. method: type: string enum: - GET - POST - PUT - PATCH description: >- The HTTP method used by this tool. Acceptable values are 'GET', 'POST', 'PUT', or 'PATCH'. Defaults to 'POST' if unspecified. id: type: string description: >- A unique identifier for the tool. Ensures distinct identification within the system. name: type: string description: >- The name of the tool, providing a human-readable identifier for display purposes. description: type: string description: >- A detailed description of the tool's purpose and functionality. Helps users understand its role within the agent. serverUrl: type: string description: >- The server URL where the tool's operations are performed. Used for network-based integrations. serverUrlSecret: type: string description: >- A secret token or key associated with the server URL. Provides secure access to the tool's backend services. createdAt: type: string description: >- The timestamp when the tool was created, formatted as an ISO 8601 string. Useful for tracking tool lifecycle. updatedAt: type: string description: >- The timestamp when the tool was last updated, formatted as an ISO 8601 string. Useful for versioning and auditing. disabled: type: boolean description: >- Indicates whether the tool is currently disabled. A disabled tool cannot be used by the agent. isVapiTool: type: boolean description: >- Indicates whether the tool is a Virtual API (VAPI) tool. VAPI tools enable interaction with external APIs. vapiId: type: string description: >- A unique identifier for the VAPI associated with this tool. Used for integration with specific API endpoints. isGlobal: type: boolean description: >- If true, this tool is globally accessible to all nodes and agents. Overrides context-specific restrictions. variablesIds: type: array items: type: string description: >- An array of variable IDs associated with the tool. Each ID maps to a specific variable used by the tool. agentId: type: string description: >- The unique identifier of the agent associated with this tool. Helps in linking the tool to a specific agent. userId: type: string description: >- The unique identifier of the user associated with this tool. Used for user-specific configurations and access control. backchannellingPhrases: type: array items: type: string description: >- An array of phrases used for backchannel communication. These phrases help in maintaining conversational flow. toolsSettings: {} fields: type: array items: type: object properties: id: type: string description: >- A unique identifier for the field, ensuring it is distinct across the system. in: type: string description: >- Defines the location or context in which the field is utilized. Common values include 'query', 'body', or 'header'. type: type: string enum: - string - number - boolean - system description: >- Specifies the variable type of the field. Acceptable values include 'string', 'number', 'boolean', and 'system'. value: description: >- The current assigned value of the field. This value is optional and supports any type depending on the field's context. defaultValue: description: >- A predefined value assigned to the field when no specific value is provided. Useful for ensuring consistent behavior. key: type: string description: >- The unique key that identifies this tool field or variable. Often used for referencing the field programmatically. description: type: string description: >- A detailed explanation of the field's purpose and usage. Helps users understand the field's role in the system. required: type: boolean description: >- Indicates whether the presence of this field is mandatory for successful operation. Defaults to false if unspecified. reusable: type: boolean description: >- Denotes whether this field can be reused across multiple contexts or tools. Useful for reducing redundancy. isEnv: type: boolean description: >- Specifies if this field represents an environment variable, typically used for configuration or deployment. isSystem: type: boolean description: >- Indicates whether this field is a system-level variable, reserved for core operations or internal use. isGlobal: type: boolean description: >- If enabled, this field remains globally accessible to the agent across all operational contexts. Useful for global constants. agentId: type: string description: >- References the unique identifier of the agent associated with this field. Helps in mapping fields to specific agents. userId: type: string description: >- References the unique identifier of the user associated with this field. Useful for user-specific customizations. required: - id additionalProperties: false channels: type: array items: type: string enum: - web-chat - whatsapp - instagram - telegram - discord - gb-chat - vapi - messenger - voice - telephony - outlook - zoho - sms description: >- The channels that the tool can be used on. required: - id - name - description additionalProperties: false internal: type: object properties: accountCreationConfig: type: string enum: - v2 websocketServer: anyOf: - type: string enum: - edge - type: string enum: - nodejs setOnResponse: type: boolean enableSetOnLoad: type: boolean voiceflowResolvedRuntime: type: string enum: - legacy - v4 additionalProperties: false disableAutoTranscribeAudio: type: boolean disableTranscriptSharing: type: boolean instaAgentUsername: type: string convoTags: type: array items: type: string vfConfig: type: object properties: alwaysEnableCardButtons: type: boolean additionalProperties: false discord: type: object properties: channelIds: type: array items: type: string webhookUrls: type: array items: type: string additionalProperties: false limits: type: object properties: maxInteractionsPerUserId: type: number enableTextIpRateLimit: type: boolean maxTextInteractionsPerIp: type: number textIpRateLimitWindowHours: type: number maxAnnualCreditsUsage: type: number maxMonthlyCreditsUsage: type: number maxMessages: type: number description: >- Maximum number of messages allowed per conversation maxMessagesRetentionDays: type: number description: Retention period in days for messages additionalProperties: false tabs: type: array items: type: object properties: key: type: string enum: - home - convos - faq label: type: string iconUrl: type: string iframeUrl: type: string iframeHeight: type: number hide: type: boolean homeSpecific: type: object properties: buttons: type: array items: type: object properties: key: type: string label: type: string iconUrl: type: string show: type: boolean iceBreakers: type: array items: type: string additionalProperties: false iceBreakers: type: array items: type: string showLiveCall: type: boolean showRecentConvo: type: boolean showDirectHandoff: type: boolean headerHeight: type: number headerTitle: type: string headerDescription: type: string required: - buttons additionalProperties: false faqSpecific: type: object properties: questions: type: array items: type: object properties: id: type: string question: type: string answer: type: string key: type: string required: - id - question - answer additionalProperties: false enableCustomFaq: type: boolean required: - questions additionalProperties: false required: - key - label additionalProperties: false kbTags: type: array items: type: string kbTagsDataMap: type: array nodes: type: array items: type: object properties: id: type: string name: type: string description: type: string description: >- A short description of what this node does and when it should be used, specially useful if the node is global or when the LLM detects automatic rerouting. instructions: type: string description: What should this LLM node do. language: type: string description: >- Preferred language code for this node responses and runtime filler/tool-status phrases (ISO-639-1, e.g. 'en', 'bg', 'es'). mcpServers: type: array items: type: object properties: name: type: string url: type: string transport: type: string enum: - sse - websocket - stdio useNodeEventSource: type: boolean reconnect: type: object properties: enabled: type: boolean maxAttempts: type: number delayMs: type: number additionalProperties: false command: type: string args: type: array items: type: string env: type: object additionalProperties: type: string cwd: type: string headers: type: object additionalProperties: type: string additionalProperties: false voiceInstructions: type: string description: >- The voice instructions for this node, will be used instead of instructions in voice channels if provided. isGlobal: type: boolean description: >- If true, this node will be always present for the LLM toolsIds: type: array items: type: string description: >- The tools that will be available for this node toolsSettings: type: object properties: googleCalendar: type: object properties: calendarId: type: string methods: type: array items: type: string enum: - create-event - check-availability - delete-event - lookup-events - update-event additionalProperties: false googleSheets: type: object properties: connectionId: type: string sheetId: type: string method: type: string sheetTabName: type: string headerRowIndex: type: number headerDescriptions: type: object additionalProperties: type: string identifierColumn: type: string identityFields: type: object properties: phone: type: string email: type: string name: type: string custom: type: string customLabel: type: string additionalProperties: false queryableFields: type: array items: type: string additionalProperties: false airtable: type: object properties: connectionId: type: string baseId: type: string tableId: type: string method: type: string fieldDescriptions: type: object additionalProperties: type: string identifierField: type: string additionalProperties: false additionalProperties: false childrenNodes: type: array items: type: object properties: nodeId: type: string condition: type: string description: >- The condition that will be checked to determine if the node should advanced to the selected node id or not or should simply loop in the current node. required: - nodeId - condition additionalProperties: false description: >- The outputs of this node which the node after finishing will choose from to either advance or to keep looping in this node till it caputures or reaches one of the mentioned routers. llmConfig: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false routerLlmConfig: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false description: >- The LLM config that will be used for the router node, NOTE: This must be a fast LLM model or the latency will be too high between responses. toolUseBias: type: number description: >- The bias of the LLM to use tools if 0 it will never use tools, if 1 it will only use tools. autoRerouter: type: object properties: enabled: type: boolean description: >- If true, the LLM will try to detect if the user has changed his mind about something enabling the AI to rewind back to the previous node. level: type: number description: >- The level of rewinding, if 1 the LLM will have access to only 1 level or rewind capabilities meaning if it went from node A > B > C > D then if it at node d & it detects it needs to be at node B again it will be able to rewind back to node B if the level is at least 2, it will not be able to detect node A unless the level is 3 or more, and so on. required: - enabled - level additionalProperties: false description: The auto rerouter of this node. type: type: string enum: - start - end - default - condition - note description: >- The type of the node, if not provided it will be a normal LLM node. rf: description: >- The position of the node in the reactflow canvas. kb: type: object properties: enabled: type: boolean maxChunks: type: number description: The max chunks to use from the KB. tags: type: array items: type: string description: >- The tags to use to filter the docs we search through smartSearch: type: boolean description: >- If true, the LLM will use the smart search to search through the docs. required: - enabled - maxChunks additionalProperties: false conditionData: type: object properties: condition: type: string description: The condition to check. required: - condition additionalProperties: false preStart: type: object properties: url: type: string description: The URL to send the GET request to. enabled: type: boolean description: >- If true, we will send a GET request to the URL before starting the node. testConvoIdValue: type: string description: The value to test the URL with. additionalProperties: false description: >- The pre start tool of the node, we will send a GET request to this URL before starting the node and provide the results to the AI. GET {url}/{user_id}/{node_name} startConfig: type: object properties: initialMessage: type: string description: >- The initial message to start the conversation with userStarts: type: boolean description: >- Whether the user or AI agent gives the first message by default for both inbound and outbound calls. inboundUserStarts: type: boolean description: >- Optional override for inbound voice calls. When set, this replaces the default first-speaker behavior for callers who dial the agent. outboundUserStarts: type: boolean description: >- Optional override for outbound voice calls. When set, this replaces the default first-speaker behavior for calls the agent places to someone else. additionalProperties: false description: Configuration for start nodes required: - id - name - description - instructions - llmConfig additionalProperties: false description: A node in the LLM chain. enableNodes: type: boolean advanced: type: object properties: serverUrl: type: string serverUrlSecret: type: string additionalProperties: false nodesSettings: type: object properties: enableUIEngineForms: type: boolean description: >- If true, the LLM will be able to show forms on web channel only. geminiLiveOptions: {} appendBeforePrompt: type: string description: >- The prompt to append to the nodes' prompt. routerLLmOptions: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false defaultLLmOptions: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false backchannelOnChoosingNodes: type: boolean description: >- If true, the LLM will say uha/i see before choosing a node. enableUiEngine: type: boolean description: >- If true, the LLM will be able to show buttons, cards, images on text channels only. fallbackModelIds: type: array items: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm silenceDetection: type: object properties: enabled: type: boolean default: false description: >- Whether to enable the silence detection feature which will insert filler phrases during long periods of silence timeoutSeconds: type: number default: 60 description: >- Number of seconds of silence before triggering a filler phrase endCallAfterNPhrases: type: number default: 1 description: >- Number of filler phrases utterances to say before ending the call if 0 it will end the call instantly after reching the timeout. additionalProperties: false startCallPhrases: type: array items: type: string fillerWordsOnToolUsage: type: boolean description: >- If true, the LLM will say filler words on tool usage, like '1 moment', 'be right back', 'you know', etc. toolUsageBackchannelPhrasesByLanguage: type: object additionalProperties: type: array items: type: string description: >- Cached filler/tool-status phrases by language code, generated on save to avoid runtime translation latency. smartEndpointing: type: object properties: enabled: type: boolean waitSeconds: type: number onPunctuationSeconds: type: number onNoPunctuationSeconds: type: number onNumberSeconds: type: number additionalProperties: false stopSpeakPlan: type: object properties: minWords: type: number description: >- The minimum words to speak from the user for the AI to be interrupted and stop talking . additionalProperties: false callTimeoutSeconds: type: number description: The timeout for the call in seconds. enableEndcallTool: type: boolean description: Whether to enable endcall tool. voiceSpecific: type: object properties: minCharacters: type: number description: >- The minimum number of characters to init the speech gen to generate audio for, the more the higher the latency will be, default is 5 maxLengthWithoutPunctuation: type: number description: >- The maximum length of the string without punctuation to init the speech gen to generate audio for, the more the higher the latency will be, default is 100 additionalProperties: false enabledGlobalTools: type: array items: type: string enum: - handoffHumanDashboardTool description: >- The global tools that will be enabled for the LLM. defaultToolOverrides: type: object additionalProperties: type: object properties: description: type: string nullable: true fields: type: object additionalProperties: type: object properties: type: type: string enum: - string - number - boolean description: type: string additionalProperties: false nullable: true additionalProperties: false description: >- Overrides for default system tools configuration. additionalProperties: false description: >- Global options for the nodes setup, text/voice. id: type: string enableFallback: type: boolean description: Enable fallback for the agent llm model. fallbackSettings: type: object properties: model: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm description: The model to use for fallback. creditThreshold: type: number description: The credit threshold for fallback. additionalProperties: false description: Fallback settings for the agent llm model. marketplace: type: object properties: enabled: type: boolean description: >- Whether to show the agent in marketplace or not. priceUSD: type: number description: >- The price for the agent, default is 0/free category: type: string enum: - Customer Support - Sales & Marketing - Productivity - Education - Entertainment - Other description: Category of the agent creatorDisplayName: type: string description: >- Custom display name for the creator in the marketplace shortDescription: type: string maxLength: 200 description: >- Short tagline/description shown in listing cards (max 200 chars) documentation: type: string description: >- Full markdown documentation explaining the agent toolsDocumentation: type: array items: type: object properties: name: type: string description: Name of the tool description: type: string description: Description of what the tool does codeUrl: type: string description: URL to the tool code/repository docsUrl: type: string description: URL to external documentation setupInstructions: type: string description: How to set up/configure the tool required: - name - description additionalProperties: false description: >- Documentation for tools included with the agent tags: type: array items: type: string description: Tags for discoverability screenshots: type: array items: type: string description: Array of screenshot URLs demoUrl: type: string description: Link to live demo videoUrl: type: string description: Link to demo/tutorial video version: type: string description: Version of the agent listing changelog: type: string description: Changelog/update history supportEmail: type: string description: Support contact email listedAt: type: number description: Unix timestamp when first listed updatedAt: type: number description: Unix timestamp of last update allowCanvasPreview: type: boolean description: >- Whether to allow users to preview the agent canvas/workflow allowTryAgent: type: boolean default: true description: >- Whether to allow users to try the agent widget likeCount: type: number default: 0 description: Total number of likes for this agent downloadCount: type: number default: 0 description: >- Total number of downloads/installs for this agent additionalProperties: false description: Marketplace settings for the agent. disableRating: type: boolean description: >- Whether to disable the chat end rating feature. chatEndMessage: type: string description: The message to show when the chat ends. aiIntroductionMessage: type: string description: >- The message to show when the user starts the chat on metachannels. enableAIIntroductionMessage: type: boolean description: >- Whether to enable the AI introduction message or not. deployed: type: boolean description: Whether the agent is deployed or not. captureIGStories: type: boolean description: Whether to capture instagram stories or not. gdprCompliance: type: boolean description: Whether to comply with GDPR or not. disableFileUpload: type: boolean buttonVariant: type: string enum: - custom - solid - bordered - light - flat - faded - shadow - ghost description: >- The variant of the button. Custom is the default original styling. inputVariant: type: string enum: - custom - flat - bordered - faded - underlined description: >- The variant of the input. Custom is the default original styling. privacyDisclaimer: type: object properties: enabled: type: boolean description: >- Whether to show privacy disclaimer before user interaction text: type: string description: >- The privacy disclaimer text to display to users privacyPolicyUrl: type: string description: URL to the privacy policy page acceptButtonLabel: type: string description: Label for the accept/proceed button privacyPolicyButtonLabel: type: string description: Label for the privacy policy link button additionalProperties: false description: >- Privacy disclaimer configuration for the agent leadCollectionRules: type: object properties: enabled: type: boolean description: >- Whether to use custom lead collection rules. If false or not set, uses default behavior (collect when email, phone_number, or phone is present) rules: type: array items: type: object properties: variables: type: array items: type: string description: >- Array of variable names that must ALL be present for this rule to match (AND condition) description: type: string description: >- Optional description of what this rule captures required: - variables additionalProperties: false description: >- Array of rule sets. A lead is collected if it matches ANY rule set (OR condition between rules). Each rule requires ALL its variables to be present (AND condition within a rule). additionalProperties: false description: >- Custom lead collection rules. Allows defining which variable combinations should trigger lead capture. If not set, defaults to collecting leads when email, phone_number, or phone is present. identityResolutionSettings: type: object properties: enabled: type: boolean description: >- Whether cross-channel identity linking is enabled. Defaults to true (on) when unset. scope: type: string enum: - agent - workspace description: >- Identity linking scope. Default agent — only link convos within this agent. softMatchEnabled: type: boolean description: >- v2 scaffold: allow confirmation-gated soft matching. Disabled by default. additionalProperties: false description: Cross-channel identity resolution settings emailConfig: type: object properties: enabled: type: boolean description: >- Whether email channel is enabled for this agent assignedEmails: type: array items: type: object properties: provider: type: string enum: - google - outlook - zoho - smtp description: >- Email provider (google, outlook, zoho, or smtp) connectionId: type: string description: >- ID of the email connection from workspace email: type: string description: Email address for display purposes replyDelay: type: number default: 0 description: >- Legacy delay in seconds before replying to an email replyDelayMinutes: type: number default: 0 description: >- Delay in minutes before AI replies. 0 means instant reply. whenToReply: type: string enum: - always - keywords - sentiment - manual default: always description: >- When to reply: always (all emails), keywords (containing specific words), sentiment (based on tone), manual (require approval) replyRule: type: string default: '' description: >- Hard rule written by the user describing when the AI should reply on this email connection. humanNotifyRule: type: string default: '' description: >- Hard rule written by the user describing when the AI should notify a human while still continuing the email reply. notifyWorkspaceOwners: type: boolean default: true description: >- Whether human escalation notifications for this email assignment should default to workspace owners. humanTakeoverEnabled: type: boolean default: false description: >- Whether AI may retake ownership of a human-owned email thread after a timeout. humanTakeoverTimeoutMinutes: type: number default: 0 description: >- Minutes to wait before AI retakes a human-owned thread. 0 disables automatic retake. replyLength: type: string enum: - short - medium - long - auto default: auto description: >- Length of replies: short (1-2 paragraphs), medium (2-4 paragraphs), long (4+ paragraphs), auto (match incoming email) replyTone: type: string enum: - professional - friendly - casual - formal - empathetic - concise - enthusiastic default: professional description: Tone of the email replies required: - provider - connectionId - email additionalProperties: false description: >- Array of configured email accounts with their specific settings additionalProperties: false description: Email channel configuration for the agent inboundEngagement: type: object properties: enabled: type: boolean channels: type: array items: type: string enum: - whatsapp - email - messenger - instagram - telegram - discord - sms - web-chat matchMode: type: string enum: - always_reply_regardless - phrases_only - ai_only - phrases_then_ai aiRule: type: string description: >- Natural-language rule that decides whether AI should reply to the inbound message. aiModelId: type: string description: >- Model used for AI-powered inbound reply decisions. triggerPhrases: type: array items: type: string description: >- Exact user-defined phrases or tags that can trigger an immediate AI reply without using the AI gate. availability: type: object properties: enabled: type: boolean timezone: type: string days: type: array items: type: string enum: - monday - tuesday - wednesday - thursday - friday - saturday - sunday startTime: type: string pattern: ^([01]\d|2[0-3]):([0-5]\d)$ endTime: type: string pattern: ^([01]\d|2[0-3]):([0-5]\d)$ additionalProperties: false applyWhen: type: string enum: - every_message - first_message_only afterOwnershipMode: type: string enum: - always_reply - continue_rules participantListMode: type: string enum: - none - allowlist - denylist participantList: type: array items: type: string description: >- Participant identifiers that should be explicitly allowed or denied before any phrase or AI rule runs. analyzeAttachmentsBeforeDecision: type: boolean enableHumanHandoffTool: type: boolean additionalProperties: false description: >- Reusable inbound engagement policy used to decide whether the AI should reply on channels like WhatsApp and email. autoTest: type: object properties: enabled: type: boolean description: >- Whether auto-testing is enabled for this agent triggerOnEdit: type: boolean description: >- Automatically run tests after agent edits testMode: type: string enum: - full - prompt-only - with-tools - with-kb - tools-only - kb-only description: Test mode configuration maxTurns: type: number description: >- Maximum number of conversation turns (undefined = natural end) naturalEnd: type: boolean description: >- Allow AI to naturally end the conversation testScenario: type: string description: >- Default test scenario/context for auto-tests enabledToolIds: type: array items: type: string description: Specific tool IDs to include in tests enableKB: type: boolean description: Whether to test knowledge base additionalProperties: false description: >- Auto-test configuration for automatic testing after agent edits langsmithConfig: type: object properties: enabled: type: boolean description: >- Whether LangSmith tracing is enabled for this agent apiKey: type: string description: >- LangSmith API Key (Service Key recommended) projectName: type: string description: >- LangSmith project name where traces will be logged endpoint: type: string description: >- LangSmith API endpoint (defaults to https://api.smith.langchain.com) additionalProperties: false description: >- LangSmith tracing configuration for this agent. If set, overrides workspace-level LangSmith settings. funnelConfig: type: object properties: enabled: type: boolean default: false steps: type: array items: type: object properties: id: type: string name: type: string description: type: string condition: type: string description: >- Natural language condition for LLM to evaluate points: type: number minimum: 0 maximum: 100 enabled: type: boolean default: true category: type: string enum: - contact_info - intent - budget - timeline - qualification - custom required: - id - name - description - condition - points additionalProperties: false default: [] notificationRules: type: array items: type: object properties: id: type: string name: type: string enabled: type: boolean default: true type: type: string enum: - score_threshold - steps_completed - data_collected scoreThreshold: type: number requiredSteps: type: array items: type: string requiredFields: type: array items: type: string cooldownStrategy: type: string enum: - immediate - once_per_threshold - once_per_lead - once_per_session recipients: type: array items: type: string description: Email addresses to notify emailTemplate: type: string requireContactInfo: type: boolean default: true description: >- Whether to require email/phone before sending notification required: - id - name - type - cooldownStrategy - recipients additionalProperties: false default: [] maxScore: type: number default: 100 evaluateOnUserMessage: type: boolean default: true evaluateOnAIMessage: type: boolean default: true additionalProperties: false description: >- AI Funnel & Lead Scoring configuration for this agent additionalProperties: false workspaceData: {} twilioAccountSid: type: string twilioAuthToken: type: string assignedGlobalPrompts: type: array items: type: string additionalProperties: false overrideInitialPrompt: type: boolean messagesHistory: type: array items: type: object properties: role: type: string enum: - user - assistant - system - tool content: type: string name: type: string timestamp: type: number tool_call_id: type: string tool_name: type: string tool_calls: {} additionalProperties: false messagesHistoryMergeMode: type: string enum: - append - replace_system_context description: >- How init messagesHistory is merged with existing lgMessages. Defaults to replace_system_context. agentOverrides: type: object properties: title: type: string description: '**Title**: The title of the agent.' description: type: string description: >- **Description**: A brief description of the agent. theme: type: string description: >- **Theme**: The visual theme of the agent, e.g., 'blue-light' or 'custom-blue-dark'. voiceConfig: type: object properties: transcriber: type: object properties: speechConfig: type: object properties: format: type: string sampleRate: type: number language: type: string additionalProperties: false modelId: type: string patienceFactor: type: number language: type: string provider: type: string enum: - deepgram - gladia - assemblyai - speechmatics - google-cloud-speech - google-live-transcription randomOptions: {} internal: type: object properties: inputAudioStream: {} debug: type: boolean additionalProperties: false apiKey: type: string platformSpecific: type: object properties: deepgram: type: object properties: keywords: type: array items: type: string description: >- Keywords to focus on in the transcription. language: type: string smart_format: type: boolean model: type: string interim_results: type: boolean endpointing: type: number no_delay: type: boolean autoLanguageDetection: type: boolean additionalProperties: false assemblyai: type: object properties: modelId: type: string required: - modelId additionalProperties: false googleCloud: type: object properties: keywords: type: array items: type: string description: >- Keywords or phrases to focus on in the transcription. model: type: string description: The Google Cloud Speech model to use. languageCode: type: string description: The language code for transcription. enableAutomaticPunctuation: type: boolean description: Enable automatic punctuation. enableWordTimeOffsets: type: boolean description: Enable word time offsets. maxAlternatives: type: number description: >- Maximum number of recognition alternatives. profanityFilter: type: boolean description: Enable profanity filter. speechContexts: type: array items: type: object properties: phrases: type: array items: type: string boost: type: number required: - phrases additionalProperties: false description: Speech contexts for better recognition. additionalProperties: false additionalProperties: false utteranceThreshold: type: number inputVoiceEnhancer: type: boolean required: - provider additionalProperties: false description: >- **Transcriber**: The configuration options for the transcriber provider used by the agent. speechGen: allOf: - type: object properties: highAudioQuality: type: boolean description: >- Generate highest quality audio possible, must have pro plan or higher on elevenlabs to enable your own key to work with this option. provider: type: string enum: - elevenlabs - deepgram - cartesia - rime-ai - openai - google-cloud - google-live - ultravox - grok-live - minimax description: Speech providers supported by the app. modelId: type: string description: The model id of the speech gen. voiceId: type: string description: The voice id for the speech service. apiKey: type: string region: type: string backgroundNoise: type: string enum: - restaurant - office - park - street punctuationBreaks: type: array items: type: string platformSpecific: type: object properties: elevenLabs: type: object properties: stability: type: number similarity_boost: type: number use_speaker_boost: type: boolean speed: type: number style: type: number additionalProperties: false playht: type: object properties: voiceId: type: string required: - voiceId additionalProperties: false cartesia: type: object properties: emotion: type: string enum: - neutral - calm - angry - content - sad - scared speed: type: number volume: type: number promptInstructions: type: string additionalProperties: false ultravox: type: object properties: temperature: type: number minimum: 0 maximum: 1 speed: type: number minimum: 0.7 maximum: 1.2 voiceOverridesProvider: type: string enum: - elevenLabs - cartesia - google - inworld - lmnt elevenLabsModel: type: string vadSettings: type: object properties: turnEndpointDelay: type: string minimumTurnDuration: type: string minimumInterruptionDuration: type: string frameActivationThreshold: type: number minimum: 0.1 maximum: 1 additionalProperties: false additionalProperties: false grokLive: type: object properties: turnDetection: type: object properties: type: type: string enum: - server_vad threshold: type: number minimum: 0.1 maximum: 0.9 silence_duration_ms: type: number minimum: 0 prefix_padding_ms: type: number minimum: 0 additionalProperties: false inputAudioTranscription: type: object properties: model: type: string default: grok-2-audio additionalProperties: false audio: type: object properties: input: type: object properties: format: type: object properties: type: type: string enum: - audio/pcm - audio/pcmu - audio/pcma rate: type: number required: - type additionalProperties: false additionalProperties: false output: type: object properties: format: type: object properties: type: type: string enum: - audio/pcm - audio/pcmu - audio/pcma rate: type: number required: - type additionalProperties: false additionalProperties: false additionalProperties: false enableWebSearch: type: boolean enableXSearch: type: boolean debug: type: boolean additionalProperties: false additionalProperties: false backChannelling: type: boolean description: >- Whether to say umm, sure, etc.. when the user is talking for too long. language: type: string enableLongMessageBackchannelling: type: boolean description: >- Whether to say umm, sure, etc.. when the user is talking for too long. backchannelMessages: type: array items: type: string backchannelInterval: type: number wordsReplacements: type: array items: type: object properties: word: type: string replacement: type: string required: - word - replacement additionalProperties: false required: - provider - type: object properties: internal: type: object properties: inputTextStream: {} audioConfig: type: object properties: format: type: string sampleRate: type: number channels: type: number language: type: string required: - format - sampleRate - channels additionalProperties: false emitOnComplete: type: boolean additionalProperties: false description: >- **Speech Generation**: The configuration options for the speech generation provider used by the agent. config: type: object properties: recordAudio: type: boolean backgroundNoise: type: string enum: - restaurant - office - street - none enableWebCalling: type: boolean firstInputChunkUNIXMs: type: number firstOutputChunkUNIXMs: type: number required: - recordAudio additionalProperties: false description: >- **Call Configuration**: The call configuration settings for the agent. additionalProperties: false light: type: boolean description: >- Whether to enable light mode for the agent. If set to true, the agent will not retain previous chat interactions for privacy reasons. enableVertex: type: boolean description: Whether to enable Vertex AI for the agent. autoOpenWidget: type: boolean description: >- Whether to auto-open the widget when the agent is loaded. createdAtUNIX: type: number description: >- The timestamp when the agent was created, formatted as an ISO 8601 string. Useful for tracking tool lifecycle. disabled: type: boolean description: >- Indicates whether the agent is disabled. If set to true, the agent will not be active. vectorDb: description: >- Specifies the vector database configuration for the agent. This database is used for handling vectorized data or embeddings. agentPlatform: type: string description: >- Specifies the platform on which the agent operates, such as web, mobile, or other environments. storageID: type: string description: >- The identifier for the agent's storage configuration. VF_DIALOGUE_API_KEY: type: string description: >- API key for VF dialogue integration, enabling the agent to communicate with external systems. VF_PROJECT_API_KEY: type: string description: >- API key associated with the VF project to authenticate and enable specific functionalities. VF_PROJECT_ID: type: string description: >- The unique identifier of the VF project linked to the agent. VF_ENVIRONMENT_ALIAS: type: string description: >- Voiceflow environment alias used for v4 session bootstrap, typically `main`. VF_KB_API_KEY: type: string description: >- API key for accessing the VF knowledge base, enabling advanced knowledge retrieval capabilities. voiceflowRuntimeMode: type: string enum: - legacy - auto - v4 description: >- Controls which Voiceflow runtime path the agent should use. roundedImageURL: type: string description: >- URL pointing to the rounded image asset used to visually represent the agent. rectangeImageURL: type: string description: >- URL pointing to the rectangular image asset for the agent's branding or display. messageDelayMS: type: number description: >- Specifies the delay in milliseconds for the agent to send messages, simulating natural conversation timing. scrollAnimation: type: boolean description: >- Indicates whether scroll animations are enabled for smooth transitions in the chat interface. proactiveMessage: type: string description: >- A predefined message the agent sends proactively to engage users. acceptFileUpload: type: boolean description: >- Determines if the agent accepts file uploads from users as part of the interaction. recordChatHistory: type: boolean description: >- Specifies whether the agent logs chat histories for future reference or analysis. chatBgURL: type: string description: >- URL of the background image used in the chat interface. disableSmoothScroll: type: boolean description: >- Disables smooth scrolling if set to true, potentially improving performance on low-end devices. isDeployed: type: boolean description: >- Indicates whether the agent is currently deployed and operational. tokensUsage: description: >- Tracks the token usage statistics for the agent, typically used in LLM integrations. maxTokensUsage: description: >- Defines the maximum allowable token usage for the agent to prevent overconsumption. fontFamily: type: string description: >- The font family used in the agent's user interface for consistent branding. branding: type: string description: >- Custom branding information or labels associated with the agent. customThemeJSONString: type: string description: >- A JSON string defining a custom theme for the agent's interface. autoStartWidget: type: boolean description: >- Determines if the widget should auto-start upon page load. allTimeTriggers: type: number description: >- The total number of triggers or interactions the agent has handled. syncBrowser: type: boolean description: >- Enables synchronization of settings or states with the browser session. delayBeforeSubmit: type: number description: >- The delay (in milliseconds) before submitting user input to the agent. messageAggregationSeconds: type: number description: >- Number of seconds the AI should wait and aggregate consecutive user messages before triggering a single reply. Applies globally to all text-based channels (Web, WhatsApp, Instagram, Messenger, Telegram, SMS, Discord). Default 0 = reply instantly after every message (current behaviour). When > 0, every new inbound message resets the wait timer (debounced); after the window of silence elapses, all buffered messages are concatenated into a single user turn and the AI is invoked once. region: type: string enum: - eu - na description: >- Defines the geographical region settings for the agent, such as localization. listenForUrlChanges: type: boolean description: >- Specifies if the agent should monitor and react to changes in the browser URL. chatForget: type: boolean description: >- If enabled, the agent will not retain previous chat interactions for privacy reasons. lang: type: string description: >- Specifies the default language for the agent's interactions and responses. enableAudioSupport: type: boolean description: >- Enables audio features, allowing the agent to process and respond to audio inputs. AITranslateTo: type: string description: >- Defines the target language for AI translation during interactions. enableAITranslate: type: boolean description: >- Activates AI-based translation for multilingual support. disableNoReplyListener: type: boolean description: >- Disables the listener for instances where no reply is detected from the agent. disableHumanHandoff: type: boolean description: >- Prevents the agent from escalating interactions to a human operator. alwaysShowHandoff: type: boolean description: >- Ensures the handoff option is always visible, regardless of agent status. manualControl: type: boolean description: >- Allows manual control over specific agent functionalities. enableVGHandoff: type: boolean description: >- Activates VG-specific handoff mechanisms for the agent. enableGeoAnalytics: type: boolean description: >- Enables geographic analytics to track user interactions based on location. hideVoiceflowAnalytics: type: boolean description: >- Hides Voiceflow-specific analytics including intents, conversation flow metrics, and rating analytics. hideCallAnalytics: type: boolean description: >- Hides voice call analytics including duration, latency, cost breakdown, call counts, and call end reasons. hideConversationMetrics: type: boolean description: >- Hides conversation metrics including user retention, conversation length, time retention, and engagement metrics. hideCustomAnalytics: type: boolean description: >- Hides user-created custom charts and metrics from the analytics dashboard. hideVapiAnalytics: type: boolean description: >- Hides VAPI-specific analytics including cost tracking and voice-related metrics. hideFunnelAnalytics: type: boolean description: >- Hides the user engagement funnel chart that tracks progression from widget views to high engagement. hideHandoffAnalytics: type: boolean description: >- Hides handoff analytics including accepted handover count, average response time, and average handling time metrics. fixedHandoffPopup: type: boolean description: >- Ensures the handoff popup remains fixed in the interface. buttonsLayout: type: string enum: - horizontal - vertical - in-footer description: >- Defines the layout and style of buttons in the agent's interface. ADVANCED_customCSS: type: string description: >- Specifies advanced custom CSS for overriding the default styling of the agent's interface. messagesLimit: type: number description: >- Sets a cap on the number of messages the agent can exchange in a single session. whatsappToken: type: string description: >- Token for authenticating and integrating with WhatsApp services. whatsappNumberId: type: string description: >- The ID associated with the WhatsApp number used by the agent. whatsappBusniessId: type: string description: >- The business ID for WhatsApp integration, enabling enterprise-level features. whatsappCustomBridge: type: object properties: enabled: type: boolean description: >- When enabled, inbound WhatsApp webhooks are forwarded to a developer-provided server and the returned payload is sent back through WhatsApp. webhookUrl: type: string description: >- Developer webhook URL that receives raw WhatsApp events and returns the outgoing WhatsApp payload to send. forwardToDeveloper: type: boolean description: >- When enabled, the raw Meta webhook payload is forwarded to the developer URL with zero alteration. The developer response is sent directly to the WhatsApp API as-is, with no normalization or processing. additionalProperties: false description: >- Developer-focused WhatsApp bridge that bypasses the built-in agent reply flow. waTestNumber: type: string description: >- A test number used during WhatsApp integration setup or testing. waVerifyPassed: type: boolean description: >- Indicates if WhatsApp verification was successfully completed. waTestPassed: type: boolean description: >- Indicates if the WhatsApp integration test was successful. twilioSmsAccountSid: type: string description: Twilio Account SID for SMS integration. twilioSmsAuthToken: type: string description: >- Twilio Auth Token for SMS integration (encrypted). twilioSmsPhoneNumber: type: string description: Twilio phone number for SMS integration. twilioSmsMode: type: string enum: - own_account - platform_rental description: >- Mode for Twilio SMS - using own account or platform rental. twilioSmsVerified: type: boolean description: >- Indicates if the Twilio SMS integration is verified. twilioSmsPhoneNumberSid: type: string description: >- Twilio IncomingPhoneNumber SID for platform rental numbers (used for releasing). twilioSmsRentalSubscriptionId: type: string description: >- Stripe subscription ID for platform rental SMS number billing. twilioSmsRentalStatus: type: string enum: - active - pending - cancelled - past_due description: Status of the SMS number rental subscription. webhookUrl: type: string description: >- The URL where the agent sends webhook events to external systems. gcloudPrivateKey: type: string description: >- Private key for Google Cloud services integration. gcloudClientEmail: type: string description: Client email for Google Cloud authentication. smartInit: type: boolean description: >- Activates smart initialization features for the agent. UIhandoffTitle: type: string description: Title displayed during UI handoff interactions. UIhandoffSubtitle: type: string description: >- Subtitle displayed during UI handoff interactions. igVerified: type: boolean description: >- Indicates if the agent's Instagram integration is verified. igAccessToken: type: string description: >- Access token for authenticating Instagram API requests. showHandoffEvenIfOffline: type: boolean description: >- Displays the handoff option even if the agent is offline. persistHandoffOnRefresh: type: boolean description: >- Maintains handoff state when users refresh or close their browser tabs. enableBugReporting: type: boolean description: >- Enables bug reporting module in the conversation's controls. blockInputDuringHandoffRequest: type: boolean description: >- Blocks chat input when a user requests human handoff but no human has been assigned yet. Only applies to VF agents. disableHandoffMessages: type: boolean description: >- When enabled, the system will not send automatic messages when handoff occurs (e.g., 'Now chatting with John'). disableHandoffStatusMessages: type: boolean description: >- When enabled, the system will not send status messages during handoff transitions (e.g., 'We are connecting you to a human', 'Handoff cancelled', 'No agent was available'). customButtonJSON_STRING: type: string description: >- JSON string defining custom buttons for the agent's interface. vg_initPrompt: type: string description: >- The initial prompt displayed or used by the VF system. vg_prompt: type: string description: >- The main prompt template used for generating responses. vg_initMessages: type: array items: type: string description: >- Initial messages loaded for the agent's conversation flow. vg_systemPrompt: type: string description: >- System-level prompt defining the agent's behavior and rules. vg_temperature: type: number description: >- Defines the temperature setting for response variability in the VF model. vg_defaultModel: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - meta-llama/llama-4-scout-17b-16e-instruct - meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm description: >- Specifies the default LLM model used by the agent for language processing. vg_maxTokens: type: number description: >- Sets the maximum token limit for generating responses. vg_kbDefaultDimension: type: number description: >- Default dimensionality for the agent's knowledge base embeddings. vg_kbTopChunks: type: number description: >- The number of top chunks retrieved from the knowledge base during a query. SECRET_API_KEY: type: string description: >- The secret API key for secure integration with VF services. vg_kbCharCount: type: number description: >- Tracks the total character count of the agent's knowledge base. vg_kbDocsNum: type: number description: >- The total number of documents stored in the agent's knowledge base. vg_enableUIEngine: type: boolean description: >- Activates the UI engine for enhanced interface capabilities. vg_enableAboutContext: type: boolean description: >- Enables contextual information about the agent for users. vg_kb_llm: type: object properties: searchMethod: anyOf: - type: string enum: - smart - type: string enum: - simple searchPrompt: type: string modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temprature: type: number maxTokens: type: number additionalProperties: false description: >- Configuration for LLM-based knowledge base queries, including search methods and model parameters. vg_enableUIEngineForms: type: boolean description: >- Enables form and input components in the UI engine for web channel. vg_max_messages_history: type: number description: >- Maximum number of messages retained in the chat history. ifKnowsThreshold: type: number description: >- Confidence threshold for determining if the agent 'knows' an answer. avatarImageUrl: type: string description: URL of the avatar image representing the agent. headerImageUrl: type: string description: >- URL of the header image displayed in the chat interface. bannerImageUrl: type: string description: >- URL of the banner image used in the agent's branding. soundEffectUrl: type: string description: >- URL of the sound effect file used for agent interactions. soundEffectEnabled: type: boolean description: Enables or disables sound effects for the agent. enableQuickFileUpload: type: boolean description: >- Allows quick file uploads during chat interactions. instagramOptions: type: object properties: pageId: type: string accessToken: type: string additionalProperties: false translateUserResponse: type: boolean starred: type: boolean flowise: type: object properties: webhookUrl: type: string webhookSecret: type: string additionalProperties: false vfSettings: type: object properties: enableIgnoreUrlPaths: type: boolean additionalProperties: false vapiConfig: type: object properties: syncKbOnSave: type: boolean PUBLIC_API_KEY: type: string PRIVATE_API_KEY: type: string enableVapiOnWeb: type: boolean overrideWithVG: type: boolean promptOnWeb: type: boolean maxCostMonthly: type: number maxMinutesMonthly: type: number useVfProject: type: boolean syncTools: type: boolean additionalProperties: false vgOptions: type: object properties: isLlmStudio: type: boolean showSources: type: boolean lightUiEngine: type: boolean legacyKb: type: boolean maxChunkSize: type: number handoffTimeout: type: number additionalProperties: false ui: type: object properties: showWhatsappButtonOnWeb: type: boolean description: >- Show little whatsapp button on web if it's connected. bgImageVisible: type: boolean widgetType: anyOf: - type: string enum: - direct-chat - type: string enum: - tabs enableFixedFeedbackBtns: type: boolean switchConversationSides: type: boolean switchDashboardOnly: type: boolean voice: type: object properties: question: type: string continueBtnLabel: type: string dismissBtnLabel: type: string disableVoiceOrb: type: boolean voiceOnlyMode: type: boolean disableChatWhenVoiceOpen: type: boolean required: - question - continueBtnLabel - dismissBtnLabel additionalProperties: false additionalProperties: false assignedToolsIds: type: array items: type: string tools: type: array items: type: object properties: isDefault: type: boolean description: >- Indicates whether this tool is the default tool for the agent. Defaults to false if unspecified. method: type: string enum: - GET - POST - PUT - PATCH description: >- The HTTP method used by this tool. Acceptable values are 'GET', 'POST', 'PUT', or 'PATCH'. Defaults to 'POST' if unspecified. id: type: string description: >- A unique identifier for the tool. Ensures distinct identification within the system. name: type: string description: >- The name of the tool, providing a human-readable identifier for display purposes. description: type: string description: >- A detailed description of the tool's purpose and functionality. Helps users understand its role within the agent. serverUrl: type: string description: >- The server URL where the tool's operations are performed. Used for network-based integrations. serverUrlSecret: type: string description: >- A secret token or key associated with the server URL. Provides secure access to the tool's backend services. createdAt: type: string description: >- The timestamp when the tool was created, formatted as an ISO 8601 string. Useful for tracking tool lifecycle. updatedAt: type: string description: >- The timestamp when the tool was last updated, formatted as an ISO 8601 string. Useful for versioning and auditing. disabled: type: boolean description: >- Indicates whether the tool is currently disabled. A disabled tool cannot be used by the agent. isVapiTool: type: boolean description: >- Indicates whether the tool is a Virtual API (VAPI) tool. VAPI tools enable interaction with external APIs. vapiId: type: string description: >- A unique identifier for the VAPI associated with this tool. Used for integration with specific API endpoints. isGlobal: type: boolean description: >- If true, this tool is globally accessible to all nodes and agents. Overrides context-specific restrictions. variablesIds: type: array items: type: string description: >- An array of variable IDs associated with the tool. Each ID maps to a specific variable used by the tool. agentId: type: string description: >- The unique identifier of the agent associated with this tool. Helps in linking the tool to a specific agent. userId: type: string description: >- The unique identifier of the user associated with this tool. Used for user-specific configurations and access control. backchannellingPhrases: type: array items: type: string description: >- An array of phrases used for backchannel communication. These phrases help in maintaining conversational flow. toolsSettings: {} fields: type: array items: type: object properties: id: type: string description: >- A unique identifier for the field, ensuring it is distinct across the system. in: type: string description: >- Defines the location or context in which the field is utilized. Common values include 'query', 'body', or 'header'. type: type: string enum: - string - number - boolean - system description: >- Specifies the variable type of the field. Acceptable values include 'string', 'number', 'boolean', and 'system'. value: description: >- The current assigned value of the field. This value is optional and supports any type depending on the field's context. defaultValue: description: >- A predefined value assigned to the field when no specific value is provided. Useful for ensuring consistent behavior. key: type: string description: >- The unique key that identifies this tool field or variable. Often used for referencing the field programmatically. description: type: string description: >- A detailed explanation of the field's purpose and usage. Helps users understand the field's role in the system. required: type: boolean description: >- Indicates whether the presence of this field is mandatory for successful operation. Defaults to false if unspecified. reusable: type: boolean description: >- Denotes whether this field can be reused across multiple contexts or tools. Useful for reducing redundancy. isEnv: type: boolean description: >- Specifies if this field represents an environment variable, typically used for configuration or deployment. isSystem: type: boolean description: >- Indicates whether this field is a system-level variable, reserved for core operations or internal use. isGlobal: type: boolean description: >- If enabled, this field remains globally accessible to the agent across all operational contexts. Useful for global constants. agentId: type: string description: >- References the unique identifier of the agent associated with this field. Helps in mapping fields to specific agents. userId: type: string description: >- References the unique identifier of the user associated with this field. Useful for user-specific customizations. required: - id additionalProperties: false channels: type: array items: type: string enum: - web-chat - whatsapp - instagram - telegram - discord - gb-chat - vapi - messenger - voice - telephony - outlook - zoho - sms description: The channels that the tool can be used on. required: - id - name - description additionalProperties: false internal: type: object properties: accountCreationConfig: type: string enum: - v2 websocketServer: anyOf: - type: string enum: - edge - type: string enum: - nodejs setOnResponse: type: boolean enableSetOnLoad: type: boolean voiceflowResolvedRuntime: type: string enum: - legacy - v4 additionalProperties: false disableAutoTranscribeAudio: type: boolean disableTranscriptSharing: type: boolean instaAgentUsername: type: string convoTags: type: array items: type: string vfConfig: type: object properties: alwaysEnableCardButtons: type: boolean additionalProperties: false discord: type: object properties: channelIds: type: array items: type: string webhookUrls: type: array items: type: string additionalProperties: false limits: type: object properties: maxInteractionsPerUserId: type: number enableTextIpRateLimit: type: boolean maxTextInteractionsPerIp: type: number textIpRateLimitWindowHours: type: number maxAnnualCreditsUsage: type: number maxMonthlyCreditsUsage: type: number maxMessages: type: number description: >- Maximum number of messages allowed per conversation maxMessagesRetentionDays: type: number description: Retention period in days for messages additionalProperties: false tabs: type: array items: type: object properties: key: type: string enum: - home - convos - faq label: type: string iconUrl: type: string iframeUrl: type: string iframeHeight: type: number hide: type: boolean homeSpecific: type: object properties: buttons: type: array items: type: object properties: key: type: string label: type: string iconUrl: type: string show: type: boolean iceBreakers: type: array items: type: string additionalProperties: false iceBreakers: type: array items: type: string showLiveCall: type: boolean showRecentConvo: type: boolean showDirectHandoff: type: boolean headerHeight: type: number headerTitle: type: string headerDescription: type: string required: - buttons additionalProperties: false faqSpecific: type: object properties: questions: type: array items: type: object properties: id: type: string question: type: string answer: type: string key: type: string required: - id - question - answer additionalProperties: false enableCustomFaq: type: boolean required: - questions additionalProperties: false required: - key - label additionalProperties: false kbTags: type: array items: type: string kbTagsDataMap: type: array nodes: type: array items: type: object properties: id: type: string name: type: string description: type: string description: >- A short description of what this node does and when it should be used, specially useful if the node is global or when the LLM detects automatic rerouting. instructions: type: string description: What should this LLM node do. language: type: string description: >- Preferred language code for this node responses and runtime filler/tool-status phrases (ISO-639-1, e.g. 'en', 'bg', 'es'). mcpServers: type: array items: type: object properties: name: type: string url: type: string transport: type: string enum: - sse - websocket - stdio useNodeEventSource: type: boolean reconnect: type: object properties: enabled: type: boolean maxAttempts: type: number delayMs: type: number additionalProperties: false command: type: string args: type: array items: type: string env: type: object additionalProperties: type: string cwd: type: string headers: type: object additionalProperties: type: string additionalProperties: false voiceInstructions: type: string description: >- The voice instructions for this node, will be used instead of instructions in voice channels if provided. isGlobal: type: boolean description: >- If true, this node will be always present for the LLM toolsIds: type: array items: type: string description: >- The tools that will be available for this node toolsSettings: type: object properties: googleCalendar: type: object properties: calendarId: type: string methods: type: array items: type: string enum: - create-event - check-availability - delete-event - lookup-events - update-event additionalProperties: false googleSheets: type: object properties: connectionId: type: string sheetId: type: string method: type: string sheetTabName: type: string headerRowIndex: type: number headerDescriptions: type: object additionalProperties: type: string identifierColumn: type: string identityFields: type: object properties: phone: type: string email: type: string name: type: string custom: type: string customLabel: type: string additionalProperties: false queryableFields: type: array items: type: string additionalProperties: false airtable: type: object properties: connectionId: type: string baseId: type: string tableId: type: string method: type: string fieldDescriptions: type: object additionalProperties: type: string identifierField: type: string additionalProperties: false additionalProperties: false childrenNodes: type: array items: type: object properties: nodeId: type: string condition: type: string description: >- The condition that will be checked to determine if the node should advanced to the selected node id or not or should simply loop in the current node. required: - nodeId - condition additionalProperties: false description: >- The outputs of this node which the node after finishing will choose from to either advance or to keep looping in this node till it caputures or reaches one of the mentioned routers. llmConfig: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false routerLlmConfig: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false description: >- The LLM config that will be used for the router node, NOTE: This must be a fast LLM model or the latency will be too high between responses. toolUseBias: type: number description: >- The bias of the LLM to use tools if 0 it will never use tools, if 1 it will only use tools. autoRerouter: type: object properties: enabled: type: boolean description: >- If true, the LLM will try to detect if the user has changed his mind about something enabling the AI to rewind back to the previous node. level: type: number description: >- The level of rewinding, if 1 the LLM will have access to only 1 level or rewind capabilities meaning if it went from node A > B > C > D then if it at node d & it detects it needs to be at node B again it will be able to rewind back to node B if the level is at least 2, it will not be able to detect node A unless the level is 3 or more, and so on. required: - enabled - level additionalProperties: false description: The auto rerouter of this node. type: type: string enum: - start - end - default - condition - note description: >- The type of the node, if not provided it will be a normal LLM node. rf: description: >- The position of the node in the reactflow canvas. kb: type: object properties: enabled: type: boolean maxChunks: type: number description: The max chunks to use from the KB. tags: type: array items: type: string description: >- The tags to use to filter the docs we search through smartSearch: type: boolean description: >- If true, the LLM will use the smart search to search through the docs. required: - enabled - maxChunks additionalProperties: false conditionData: type: object properties: condition: type: string description: The condition to check. required: - condition additionalProperties: false preStart: type: object properties: url: type: string description: The URL to send the GET request to. enabled: type: boolean description: >- If true, we will send a GET request to the URL before starting the node. testConvoIdValue: type: string description: The value to test the URL with. additionalProperties: false description: >- The pre start tool of the node, we will send a GET request to this URL before starting the node and provide the results to the AI. GET {url}/{user_id}/{node_name} startConfig: type: object properties: initialMessage: type: string description: >- The initial message to start the conversation with userStarts: type: boolean description: >- Whether the user or AI agent gives the first message by default for both inbound and outbound calls. inboundUserStarts: type: boolean description: >- Optional override for inbound voice calls. When set, this replaces the default first-speaker behavior for callers who dial the agent. outboundUserStarts: type: boolean description: >- Optional override for outbound voice calls. When set, this replaces the default first-speaker behavior for calls the agent places to someone else. additionalProperties: false description: Configuration for start nodes required: - id - name - description - instructions - llmConfig additionalProperties: false description: A node in the LLM chain. enableNodes: type: boolean advanced: type: object properties: serverUrl: type: string serverUrlSecret: type: string additionalProperties: false nodesSettings: type: object properties: enableUIEngineForms: type: boolean description: >- If true, the LLM will be able to show forms on web channel only. geminiLiveOptions: {} appendBeforePrompt: type: string description: The prompt to append to the nodes' prompt. routerLLmOptions: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false defaultLLmOptions: type: object properties: modelId: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm temperature: type: number description: The temperature of the LLM maxTokens: type: number description: The max tokens of the LLM serviceTier: type: string enum: - default - priority customModelId: type: string description: The custom model id of the LLM serverUrl: type: string description: The server url of the LLM apiKey: type: string description: The api key of the LLM required: - modelId - temperature - maxTokens additionalProperties: false backchannelOnChoosingNodes: type: boolean description: >- If true, the LLM will say uha/i see before choosing a node. enableUiEngine: type: boolean description: >- If true, the LLM will be able to show buttons, cards, images on text channels only. fallbackModelIds: type: array items: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - >- meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - >- meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm silenceDetection: type: object properties: enabled: type: boolean default: false description: >- Whether to enable the silence detection feature which will insert filler phrases during long periods of silence timeoutSeconds: type: number default: 60 description: >- Number of seconds of silence before triggering a filler phrase endCallAfterNPhrases: type: number default: 1 description: >- Number of filler phrases utterances to say before ending the call if 0 it will end the call instantly after reching the timeout. additionalProperties: false startCallPhrases: type: array items: type: string fillerWordsOnToolUsage: type: boolean description: >- If true, the LLM will say filler words on tool usage, like '1 moment', 'be right back', 'you know', etc. toolUsageBackchannelPhrasesByLanguage: type: object additionalProperties: type: array items: type: string description: >- Cached filler/tool-status phrases by language code, generated on save to avoid runtime translation latency. smartEndpointing: type: object properties: enabled: type: boolean waitSeconds: type: number onPunctuationSeconds: type: number onNoPunctuationSeconds: type: number onNumberSeconds: type: number additionalProperties: false stopSpeakPlan: type: object properties: minWords: type: number description: >- The minimum words to speak from the user for the AI to be interrupted and stop talking . additionalProperties: false callTimeoutSeconds: type: number description: The timeout for the call in seconds. enableEndcallTool: type: boolean description: Whether to enable endcall tool. voiceSpecific: type: object properties: minCharacters: type: number description: >- The minimum number of characters to init the speech gen to generate audio for, the more the higher the latency will be, default is 5 maxLengthWithoutPunctuation: type: number description: >- The maximum length of the string without punctuation to init the speech gen to generate audio for, the more the higher the latency will be, default is 100 additionalProperties: false enabledGlobalTools: type: array items: type: string enum: - handoffHumanDashboardTool description: >- The global tools that will be enabled for the LLM. defaultToolOverrides: type: object additionalProperties: type: object properties: description: type: string nullable: true fields: type: object additionalProperties: type: object properties: type: type: string enum: - string - number - boolean description: type: string additionalProperties: false nullable: true additionalProperties: false description: >- Overrides for default system tools configuration. additionalProperties: false description: Global options for the nodes setup, text/voice. enableFallback: type: boolean description: Enable fallback for the agent llm model. fallbackSettings: type: object properties: model: type: string enum: - moonshotai/Kimi-K2-Instruct - moonshotai/Kimi-K2.5 - moonshotai/Kimi-K2-Thinking - moonshotai/Kimi-K2.5-fast - zai-org/GLM-5 - zai-org/GLM-4.7-FP8 - zai-org/GLM-4.5 - zai-org/GLM-4.5-Air - MiniMaxAI/MiniMax-M2.1 - MiniMaxAI/MiniMax-M2.5 - deepseek-ai/DeepSeek-V3.2 - deepseek-ai/DeepSeek-V4-Pro - deepseek-ai/DeepSeek-V3.2-fast - deepseek-ai/DeepSeek-R1-0528 - deepseek-ai/DeepSeek-R1-0528-fast - deepseek-ai/DeepSeek-V3-0324 - deepseek-ai/DeepSeek-V3-0324-fast - Qwen/Qwen3.5-397B-A17B - Qwen/Qwen3-235B-A22B-Instruct-2507 - Qwen/Qwen3-235B-A22B-Thinking-2507 - Qwen/Qwen3-30B-A3B-Instruct-2507 - Qwen/Qwen3-30B-A3B-Thinking-2507 - Qwen/Qwen3-Coder-480B-A35B-Instruct - Qwen/Qwen3-Coder-30B-A3B-Instruct - Qwen/Qwen3-Next-80B-A3B-Thinking - Qwen/Qwen3-32B - Qwen/Qwen3-32B-fast - Qwen/Qwen2.5-Coder-7B-fast - meta-llama/Llama-3.3-70B-Instruct-fast - meta-llama/Llama-3.3-70B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct - meta-llama/Meta-Llama-3.1-8B-Instruct-fast - nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 - nvidia/nemotron-3-super-120b-a12b - nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B - google/gemma-3-27b-it - google/gemma-3-27b-it-fast - google/gemma-2-2b-it - google/gemma-2-9b-it-fast - openai/gpt-oss-120b - openai/gpt-oss-20b - NousResearch/Hermes-4-405B - NousResearch/Hermes-4-70B - PrimeIntellect/INTELLECT-3 - gpt-5.4-thinking-latest - gpt-5.4 - gpt-5.3-chat-latest - gpt-5.2-2025-12-11 - gpt-5-chat-latest - gpt-5-2025-08-07 - gpt-5-mini-2025-08-07 - gpt-5-nano-2025-08-07 - o1 - o1-mini - o3-mini - azure-eu-gpt-4o - azure-na-gpt-4o - gpt-4o - gpt-4o-mini - gpt-4.5-preview-2025-02-27 - gpt-4.1-2025-04-14 - gpt-4.1-mini-2025-04-14 - >- ft:gpt-4o-mini-2024-07-18:personal:4o-with-tools-t11:A6mByttv - gpt-3.5-turbo - grok-3-fast - grok-3-mini - grok-2-latest - claude-opus-4-7 - claude-opus-4-6 - claude-opus-4-5-20251101 - claude-opus-4-20250514 - claude-sonnet-4-5-20250929 - claude-haiku-4-5-20251001 - claude-sonnet-4-20250514 - claude-3-7-sonnet-20250219 - claude-3-5-sonnet-20241022 - claude-3-5-haiku-20241022 - claude-3-opus-20240229 - claude-3-sonnet-20240229 - deepseek-chat - gemini-3.5-flash - gemini-3.1-pro-preview - gemini-3.1-flash-lite - gemini-3.1-flash-lite-preview - gemini-3-pro-preview - gemini-3-flash-preview - gemini-2.5-pro - gemini-2.5-pro-preview-03-25 - gemini-2.5-pro-exp-03-25 - gemini-2.5-flash - gemini-2.5-flash-preview-05-20 - gemini-2.0-flash-thinking-exp-1219 - gemini-2.0-flash-exp - gemini-1.5-pro - gemini-1.5-flash - gemini-1.0-pro - deepseek-r1-distill-llama-70b - meta-llama/llama-4-scout-17b-16e-instruct - >- meta-llama/llama-4-maverick-17b-128e-instruct - llama-3.3-70b-versatile - llama-3.1-8b-instant - llama3-70b-8192 - gemma2-9b-it - gemma-4-31b-dense - gemma-4-26b-moe - gemma-7b-it - qwen-max-latest - qwen-plus-latest - qwen-turbo-latest - custom-llm description: The model to use for fallback. creditThreshold: type: number description: The credit threshold for fallback. additionalProperties: false description: Fallback settings for the agent llm model. marketplace: type: object properties: enabled: type: boolean description: >- Whether to show the agent in marketplace or not. priceUSD: type: number description: The price for the agent, default is 0/free category: type: string enum: - Customer Support - Sales & Marketing - Productivity - Education - Entertainment - Other description: Category of the agent creatorDisplayName: type: string description: >- Custom display name for the creator in the marketplace shortDescription: type: string maxLength: 200 description: >- Short tagline/description shown in listing cards (max 200 chars) documentation: type: string description: >- Full markdown documentation explaining the agent toolsDocumentation: type: array items: type: object properties: name: type: string description: Name of the tool description: type: string description: Description of what the tool does codeUrl: type: string description: URL to the tool code/repository docsUrl: type: string description: URL to external documentation setupInstructions: type: string description: How to set up/configure the tool required: - name - description additionalProperties: false description: >- Documentation for tools included with the agent tags: type: array items: type: string description: Tags for discoverability screenshots: type: array items: type: string description: Array of screenshot URLs demoUrl: type: string description: Link to live demo videoUrl: type: string description: Link to demo/tutorial video version: type: string description: Version of the agent listing changelog: type: string description: Changelog/update history supportEmail: type: string description: Support contact email listedAt: type: number description: Unix timestamp when first listed updatedAt: type: number description: Unix timestamp of last update allowCanvasPreview: type: boolean description: >- Whether to allow users to preview the agent canvas/workflow allowTryAgent: type: boolean default: true description: >- Whether to allow users to try the agent widget likeCount: type: number default: 0 description: Total number of likes for this agent downloadCount: type: number default: 0 description: >- Total number of downloads/installs for this agent additionalProperties: false description: Marketplace settings for the agent. disableRating: type: boolean description: Whether to disable the chat end rating feature. chatEndMessage: type: string description: The message to show when the chat ends. aiIntroductionMessage: type: string description: >- The message to show when the user starts the chat on metachannels. enableAIIntroductionMessage: type: boolean description: >- Whether to enable the AI introduction message or not. deployed: type: boolean description: Whether the agent is deployed or not. captureIGStories: type: boolean description: Whether to capture instagram stories or not. gdprCompliance: type: boolean description: Whether to comply with GDPR or not. disableFileUpload: type: boolean buttonVariant: type: string enum: - custom - solid - bordered - light - flat - faded - shadow - ghost description: >- The variant of the button. Custom is the default original styling. inputVariant: type: string enum: - custom - flat - bordered - faded - underlined description: >- The variant of the input. Custom is the default original styling. privacyDisclaimer: type: object properties: enabled: type: boolean description: >- Whether to show privacy disclaimer before user interaction text: type: string description: >- The privacy disclaimer text to display to users privacyPolicyUrl: type: string description: URL to the privacy policy page acceptButtonLabel: type: string description: Label for the accept/proceed button privacyPolicyButtonLabel: type: string description: Label for the privacy policy link button additionalProperties: false description: Privacy disclaimer configuration for the agent leadCollectionRules: type: object properties: enabled: type: boolean description: >- Whether to use custom lead collection rules. If false or not set, uses default behavior (collect when email, phone_number, or phone is present) rules: type: array items: type: object properties: variables: type: array items: type: string description: >- Array of variable names that must ALL be present for this rule to match (AND condition) description: type: string description: >- Optional description of what this rule captures required: - variables additionalProperties: false description: >- Array of rule sets. A lead is collected if it matches ANY rule set (OR condition between rules). Each rule requires ALL its variables to be present (AND condition within a rule). additionalProperties: false description: >- Custom lead collection rules. Allows defining which variable combinations should trigger lead capture. If not set, defaults to collecting leads when email, phone_number, or phone is present. identityResolutionSettings: type: object properties: enabled: type: boolean description: >- Whether cross-channel identity linking is enabled. Defaults to true (on) when unset. scope: type: string enum: - agent - workspace description: >- Identity linking scope. Default agent — only link convos within this agent. softMatchEnabled: type: boolean description: >- v2 scaffold: allow confirmation-gated soft matching. Disabled by default. additionalProperties: false description: Cross-channel identity resolution settings emailConfig: type: object properties: enabled: type: boolean description: >- Whether email channel is enabled for this agent assignedEmails: type: array items: type: object properties: provider: type: string enum: - google - outlook - zoho - smtp description: >- Email provider (google, outlook, zoho, or smtp) connectionId: type: string description: >- ID of the email connection from workspace email: type: string description: Email address for display purposes replyDelay: type: number default: 0 description: >- Legacy delay in seconds before replying to an email replyDelayMinutes: type: number default: 0 description: >- Delay in minutes before AI replies. 0 means instant reply. whenToReply: type: string enum: - always - keywords - sentiment - manual default: always description: >- When to reply: always (all emails), keywords (containing specific words), sentiment (based on tone), manual (require approval) replyRule: type: string default: '' description: >- Hard rule written by the user describing when the AI should reply on this email connection. humanNotifyRule: type: string default: '' description: >- Hard rule written by the user describing when the AI should notify a human while still continuing the email reply. notifyWorkspaceOwners: type: boolean default: true description: >- Whether human escalation notifications for this email assignment should default to workspace owners. humanTakeoverEnabled: type: boolean default: false description: >- Whether AI may retake ownership of a human-owned email thread after a timeout. humanTakeoverTimeoutMinutes: type: number default: 0 description: >- Minutes to wait before AI retakes a human-owned thread. 0 disables automatic retake. replyLength: type: string enum: - short - medium - long - auto default: auto description: >- Length of replies: short (1-2 paragraphs), medium (2-4 paragraphs), long (4+ paragraphs), auto (match incoming email) replyTone: type: string enum: - professional - friendly - casual - formal - empathetic - concise - enthusiastic default: professional description: Tone of the email replies required: - provider - connectionId - email additionalProperties: false description: >- Array of configured email accounts with their specific settings additionalProperties: false description: Email channel configuration for the agent inboundEngagement: type: object properties: enabled: type: boolean channels: type: array items: type: string enum: - whatsapp - email - messenger - instagram - telegram - discord - sms - web-chat matchMode: type: string enum: - always_reply_regardless - phrases_only - ai_only - phrases_then_ai aiRule: type: string description: >- Natural-language rule that decides whether AI should reply to the inbound message. aiModelId: type: string description: >- Model used for AI-powered inbound reply decisions. triggerPhrases: type: array items: type: string description: >- Exact user-defined phrases or tags that can trigger an immediate AI reply without using the AI gate. availability: type: object properties: enabled: type: boolean timezone: type: string days: type: array items: type: string enum: - monday - tuesday - wednesday - thursday - friday - saturday - sunday startTime: type: string pattern: ^([01]\d|2[0-3]):([0-5]\d)$ endTime: type: string pattern: ^([01]\d|2[0-3]):([0-5]\d)$ additionalProperties: false applyWhen: type: string enum: - every_message - first_message_only afterOwnershipMode: type: string enum: - always_reply - continue_rules participantListMode: type: string enum: - none - allowlist - denylist participantList: type: array items: type: string description: >- Participant identifiers that should be explicitly allowed or denied before any phrase or AI rule runs. analyzeAttachmentsBeforeDecision: type: boolean enableHumanHandoffTool: type: boolean additionalProperties: false description: >- Reusable inbound engagement policy used to decide whether the AI should reply on channels like WhatsApp and email. autoTest: type: object properties: enabled: type: boolean description: >- Whether auto-testing is enabled for this agent triggerOnEdit: type: boolean description: Automatically run tests after agent edits testMode: type: string enum: - full - prompt-only - with-tools - with-kb - tools-only - kb-only description: Test mode configuration maxTurns: type: number description: >- Maximum number of conversation turns (undefined = natural end) naturalEnd: type: boolean description: Allow AI to naturally end the conversation testScenario: type: string description: Default test scenario/context for auto-tests enabledToolIds: type: array items: type: string description: Specific tool IDs to include in tests enableKB: type: boolean description: Whether to test knowledge base additionalProperties: false description: >- Auto-test configuration for automatic testing after agent edits langsmithConfig: type: object properties: enabled: type: boolean description: >- Whether LangSmith tracing is enabled for this agent apiKey: type: string description: LangSmith API Key (Service Key recommended) projectName: type: string description: >- LangSmith project name where traces will be logged endpoint: type: string description: >- LangSmith API endpoint (defaults to https://api.smith.langchain.com) additionalProperties: false description: >- LangSmith tracing configuration for this agent. If set, overrides workspace-level LangSmith settings. funnelConfig: type: object properties: enabled: type: boolean default: false steps: type: array items: type: object properties: id: type: string name: type: string description: type: string condition: type: string description: >- Natural language condition for LLM to evaluate points: type: number minimum: 0 maximum: 100 enabled: type: boolean default: true category: type: string enum: - contact_info - intent - budget - timeline - qualification - custom required: - id - name - description - condition - points additionalProperties: false default: [] notificationRules: type: array items: type: object properties: id: type: string name: type: string enabled: type: boolean default: true type: type: string enum: - score_threshold - steps_completed - data_collected scoreThreshold: type: number requiredSteps: type: array items: type: string requiredFields: type: array items: type: string cooldownStrategy: type: string enum: - immediate - once_per_threshold - once_per_lead - once_per_session recipients: type: array items: type: string description: Email addresses to notify emailTemplate: type: string requireContactInfo: type: boolean default: true description: >- Whether to require email/phone before sending notification required: - id - name - type - cooldownStrategy - recipients additionalProperties: false default: [] maxScore: type: number default: 100 evaluateOnUserMessage: type: boolean default: true evaluateOnAIMessage: type: boolean default: true additionalProperties: false description: >- AI Funnel & Lead Scoring configuration for this agent additionalProperties: false description: This is the overrides to use for the agent. toolsOverrides: type: array items: type: object properties: isDefault: type: boolean description: >- Indicates whether this tool is the default tool for the agent. Defaults to false if unspecified. method: type: string enum: - GET - POST - PUT - PATCH description: >- The HTTP method used by this tool. Acceptable values are 'GET', 'POST', 'PUT', or 'PATCH'. Defaults to 'POST' if unspecified. name: type: string description: >- The name of the tool, providing a human-readable identifier for display purposes. description: type: string description: >- A detailed description of the tool's purpose and functionality. Helps users understand its role within the agent. serverUrl: type: string description: >- The server URL where the tool's operations are performed. Used for network-based integrations. serverUrlSecret: type: string description: >- A secret token or key associated with the server URL. Provides secure access to the tool's backend services. disabled: type: boolean description: >- Indicates whether the tool is currently disabled. A disabled tool cannot be used by the agent. isVapiTool: type: boolean description: >- Indicates whether the tool is a Virtual API (VAPI) tool. VAPI tools enable interaction with external APIs. vapiId: type: string description: >- A unique identifier for the VAPI associated with this tool. Used for integration with specific API endpoints. isGlobal: type: boolean description: >- If true, this tool is globally accessible to all nodes and agents. Overrides context-specific restrictions. backchannellingPhrases: type: array items: type: string description: >- An array of phrases used for backchannel communication. These phrases help in maintaining conversational flow. toolsSettings: {} fields: type: array items: type: object properties: id: type: string description: >- A unique identifier for the field, ensuring it is distinct across the system. in: type: string description: >- Defines the location or context in which the field is utilized. Common values include 'query', 'body', or 'header'. type: type: string enum: - string - number - boolean - system description: >- Specifies the variable type of the field. Acceptable values include 'string', 'number', 'boolean', and 'system'. value: description: >- The current assigned value of the field. This value is optional and supports any type depending on the field's context. defaultValue: description: >- A predefined value assigned to the field when no specific value is provided. Useful for ensuring consistent behavior. key: type: string description: >- The unique key that identifies this tool field or variable. Often used for referencing the field programmatically. description: type: string description: >- A detailed explanation of the field's purpose and usage. Helps users understand the field's role in the system. required: type: boolean description: >- Indicates whether the presence of this field is mandatory for successful operation. Defaults to false if unspecified. reusable: type: boolean description: >- Denotes whether this field can be reused across multiple contexts or tools. Useful for reducing redundancy. isEnv: type: boolean description: >- Specifies if this field represents an environment variable, typically used for configuration or deployment. isSystem: type: boolean description: >- Indicates whether this field is a system-level variable, reserved for core operations or internal use. isGlobal: type: boolean description: >- If enabled, this field remains globally accessible to the agent across all operational contexts. Useful for global constants. agentId: type: string description: >- References the unique identifier of the agent associated with this field. Helps in mapping fields to specific agents. userId: type: string description: >- References the unique identifier of the user associated with this field. Useful for user-specific customizations. required: - id additionalProperties: false channels: type: array items: type: string enum: - web-chat - whatsapp - instagram - telegram - discord - gb-chat - vapi - messenger - voice - telephony - outlook - zoho - sms description: The channels that the tool can be used on. required: - name - description additionalProperties: false description: >- This will create new tools for the agent or override existing ones to be used for the call/interaction. variablesOverrides: type: array items: type: object properties: in: type: string description: >- Defines the location or context in which the field is utilized. Common values include 'query', 'body', or 'header'. type: type: string enum: - string - number - boolean - system description: >- Specifies the variable type of the field. Acceptable values include 'string', 'number', 'boolean', and 'system'. value: description: >- The current assigned value of the field. This value is optional and supports any type depending on the field's context. defaultValue: description: >- A predefined value assigned to the field when no specific value is provided. Useful for ensuring consistent behavior. key: type: string description: >- The unique key that identifies this tool field or variable. Often used for referencing the field programmatically. description: type: string description: >- A detailed explanation of the field's purpose and usage. Helps users understand the field's role in the system. required: type: boolean description: >- Indicates whether the presence of this field is mandatory for successful operation. Defaults to false if unspecified. reusable: type: boolean description: >- Denotes whether this field can be reused across multiple contexts or tools. Useful for reducing redundancy. isEnv: type: boolean description: >- Specifies if this field represents an environment variable, typically used for configuration or deployment. isSystem: type: boolean description: >- Indicates whether this field is a system-level variable, reserved for core operations or internal use. isGlobal: type: boolean description: >- If enabled, this field remains globally accessible to the agent across all operational contexts. Useful for global constants. additionalProperties: false description: >- This will create new variables for the agent or override existing ones to be used for the call/interaction, if the variable is env or has a default value the agent will be able to know about it and use it globally on any node. callerInfo: {} callInfo: {} currentNode: type: string description: Current node to resume the call from. awaitListenUrl: type: boolean description: >- If true, the call will wait for the listen url to be generated before starting the call. additionalProperties: false description: >- Agent runtime overrides (tools, variables, messagesHistory, etc.). additionalProperties: false description: >- Advanced options for outbound SMS delivery and conversation handling. required: - agentId - to additionalProperties: false responses: '200': description: Successful response content: application/json: schema: type: object properties: success: type: boolean status: type: string dryRun: type: boolean skipped: type: boolean reason: type: string agentId: type: string to: type: string from: type: string convoId: type: string mode: type: string enum: - direct - agent senderMode: type: string enum: - own_account - platform_rental campaignId: type: string messageSid: type: string messageSids: type: array items: type: string messageCount: type: number mediaCount: type: number messagesPreview: type: array items: type: string segments: type: array items: type: object properties: messageSid: type: string status: type: string segmentIndex: type: number required: - messageSid - segmentIndex additionalProperties: false required: - success additionalProperties: false default: $ref: '#/components/responses/error' security: - Authorization: [] components: responses: error: description: Error response content: application/json: schema: type: object properties: message: type: string code: type: string issues: type: array items: type: object properties: message: type: string required: - message additionalProperties: false required: - message - code additionalProperties: false securitySchemes: Authorization: type: http scheme: bearer ````