> This is a page from the ElevenLabs documentation. For a complete page index, fetch https://elevenlabs.io/docs/llms.txt. For the full documentation in a single file, fetch https://elevenlabs.io/docs/llms-full.txt. # Create tool POST https://api.elevenlabs.io/v1/convai/tools Content-Type: application/json Add a new tool to the available tools in the workspace. Reference: https://elevenlabs.io/docs/api-reference/tools/create ## OpenAPI Specification ```yaml openapi: 3.1.0 info: title: api version: 1.0.0 paths: /v1/convai/tools: post: operationId: create summary: Add Tool description: Add a new tool to the available tools in the workspace. tags: - subpackage_conversationalAi.subpackage_conversationalAi/tools parameters: - name: xi-api-key in: header required: false schema: type: string responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/ToolResponseModel' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' requestBody: content: application/json: schema: $ref: '#/components/schemas/ToolRequestModel' servers: - url: https://api.elevenlabs.io - url: https://api.us.elevenlabs.io - url: https://api.eu.residency.elevenlabs.io - url: https://api.in.residency.elevenlabs.io components: schemas: PreToolSpeechMode: type: string enum: - auto - force - 'off' default: auto title: PreToolSpeechMode DynamicVariableAssignment: type: object properties: source: type: string enum: - response default: response description: >- The source to extract the value from. Currently only 'response' is supported. dynamic_variable: type: string description: The name of the dynamic variable to assign the extracted value to value_path: type: string description: >- Dot notation path to extract the value from the source (e.g., 'user.name' or 'data.0.id') sanitize: type: boolean default: false description: >- If true, this assignment's value will be removed from the tool response before sending to the LLM and transcript, but still processed for variable assignment. preserve_native_type: type: boolean default: false description: >- If true, non-scalar values (lists, objects) extracted from the tool response are stored as their native type instead of being stringified to JSON. Enable this to use extracted arrays directly as list dynamic variables. required: - dynamic_variable - value_path description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables. title: DynamicVariableAssignment ToolCallSoundType: type: string enum: - typing - elevator1 - elevator2 - elevator3 - elevator4 description: Predefined tool call sound types. title: ToolCallSoundType ToolCallSoundBehavior: type: string enum: - auto - always default: auto description: Determines how the tool call sound should be played. title: ToolCallSoundBehavior ToolErrorHandlingMode: type: string enum: - auto - summarized - passthrough - hide default: auto description: >- Controls how tool errors are processed before being shared with the agent. title: ToolErrorHandlingMode LiteralJsonSchemaPropertyType: type: string enum: - boolean - string - integer - number title: LiteralJsonSchemaPropertyType LiteralJsonSchemaPropertyConstantValue: oneOf: - type: string - type: integer - type: number format: double - type: boolean description: >- A constant value to use for this property. Mutually exclusive with description, dynamic_variable, is_system_provided, and is_omitted. title: LiteralJsonSchemaPropertyConstantValue LiteralJsonSchemaProperty: type: object properties: type: $ref: '#/components/schemas/LiteralJsonSchemaPropertyType' description: type: string default: '' description: >- The description of the property. When set, the LLM will provide the value based on this description. Mutually exclusive with dynamic_variable, is_system_provided, constant_value, and is_omitted. enum: type: - array - 'null' items: type: string description: List of allowed string values for string type parameters is_system_provided: type: boolean default: false description: >- If true, the value will be populated by the system at runtime. Used by API Integration Webhook tools for templating. Mutually exclusive with description, dynamic_variable, constant_value, and is_omitted. dynamic_variable: type: string default: '' description: >- The name of the dynamic variable to use for this property's value. Mutually exclusive with description, is_system_provided, constant_value, and is_omitted. constant_value: $ref: '#/components/schemas/LiteralJsonSchemaPropertyConstantValue' default: '' description: >- A constant value to use for this property. Mutually exclusive with description, dynamic_variable, is_system_provided, and is_omitted. is_omitted: type: boolean default: false description: >- If true, this parameter will be completely omitted from the request. Only valid for optional parameters. Mutually exclusive with description, dynamic_variable, is_system_provided, and constant_value. required: - type description: >- Schema property for literal JSON types. IMPORTANT: Only ONE of the following fields can be set: description (LLM provides value), dynamic_variable (value from variable), is_system_provided (system provides value), constant_value (fixed value), or is_omitted (parameter is omitted). These are mutually exclusive. title: LiteralJsonSchemaProperty ArrayJsonSchemaPropertyInputItems: oneOf: - $ref: '#/components/schemas/LiteralJsonSchemaProperty' - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Input' - $ref: '#/components/schemas/ArrayJsonSchemaProperty-Input' description: Schema for array elements. title: ArrayJsonSchemaPropertyInputItems ArrayJsonSchemaPropertyInputConstantValueItems: oneOf: - type: string - type: integer - type: number format: double - type: boolean title: ArrayJsonSchemaPropertyInputConstantValueItems ArrayJsonSchemaProperty-Input: type: object properties: type: type: string enum: - array default: array description: type: string default: '' items: $ref: '#/components/schemas/ArrayJsonSchemaPropertyInputItems' default: type: string description: Array element is_system_provided: false dynamic_variable: '' constant_value: '' is_omitted: false description: Schema for array elements. dynamic_variable: type: string default: '' description: >- When set, the entire array is populated from this dynamic variable at runtime. Mutually exclusive with description (LLM-provided array) and constant_value. constant_value: type: - array - 'null' items: $ref: >- #/components/schemas/ArrayJsonSchemaPropertyInputConstantValueItems description: >- When set, the entire array uses this constant value at runtime. Mutually exclusive with description (LLM-provided array) and dynamic_variable. title: ArrayJsonSchemaProperty-Input ObjectJsonSchemaPropertyInput: oneOf: - $ref: '#/components/schemas/LiteralJsonSchemaProperty' - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Input' - $ref: '#/components/schemas/ArrayJsonSchemaProperty-Input' title: ObjectJsonSchemaPropertyInput RequiredConstraint: type: object properties: required: type: array items: type: string required: - required description: A set of fields that must all be present to satisfy this constraint. title: RequiredConstraint RequiredConstraints: type: object properties: any_of: type: array items: $ref: '#/components/schemas/RequiredConstraint' all_of: type: array items: $ref: '#/components/schemas/RequiredConstraint' description: >- Wrapper for anyOf/allOf composition constraints scoped to required fields. title: RequiredConstraints ObjectJsonSchemaProperty-Input: type: object properties: type: type: string enum: - object default: object required: type: array items: type: string description: type: string default: '' properties: type: object additionalProperties: $ref: '#/components/schemas/ObjectJsonSchemaPropertyInput' required_constraints: oneOf: - $ref: '#/components/schemas/RequiredConstraints' - type: 'null' title: ObjectJsonSchemaProperty-Input DynamicVariableNestedValueType-Input: description: Any type title: DynamicVariableNestedValueType-Input DynamicVariableValueTypeInput4: type: array items: $ref: '#/components/schemas/DynamicVariableNestedValueType-Input' title: DynamicVariableValueTypeInput4 DynamicVariableValueTypeInput: oneOf: - type: string - type: number format: double - type: integer - type: boolean - $ref: '#/components/schemas/DynamicVariableValueTypeInput4' title: DynamicVariableValueTypeInput DynamicVariableValueType-Input: oneOf: - $ref: '#/components/schemas/DynamicVariableValueTypeInput' - type: 'null' title: DynamicVariableValueType-Input DynamicVariablesConfig-Input: type: object properties: dynamic_variable_placeholders: type: object additionalProperties: $ref: '#/components/schemas/DynamicVariableValueType-Input' description: A dictionary of dynamic variable placeholders and their values title: DynamicVariablesConfig-Input ToolExecutionMode: type: string enum: - immediate - post_tool_speech - async default: immediate title: ToolExecutionMode ProcedureType: type: string enum: - free_form - deterministic default: free_form title: ProcedureType GuardrailExecutionMode: type: string enum: - streaming - blocking default: streaming title: GuardrailExecutionMode CustomGuardrailConfigTriggerAction: oneOf: - type: object properties: type: type: string enum: - end_call default: end_call required: - type description: EndCallTriggerAction variant - type: object properties: type: type: string enum: - retry default: retry feedback: type: string default: >- Your response was blocked by a guardrail that blocks content that matches this condition/category: '{{trigger_reason}}' During your next turn you must tell the user "I'm sorry but I can't answer that question, would you like to know something else?". description: >- Custom feedback to inject into the agent when retrying after guardrail trigger. required: - type description: RetryTriggerAction variant discriminator: propertyName: type title: CustomGuardrailConfigTriggerAction CustomGuardrailConfig: type: object properties: is_enabled: type: boolean default: false name: type: string description: User-facing name for this guardrail prompt: type: string description: >- Instruction describing what to block, e.g. 'don't talk about politics' execution_mode: $ref: '#/components/schemas/GuardrailExecutionMode' default: streaming trigger_action: $ref: '#/components/schemas/CustomGuardrailConfigTriggerAction' required: - name - prompt description: Single custom guardrail configuration title: CustomGuardrailConfig ProcedureAtVersion-Input: type: object properties: procedure_id: type: string description: Procedure ID name: type: string description: Procedure name content: type: string description: Procedure content type: $ref: '#/components/schemas/ProcedureType' default: free_form guardrails: type: array items: $ref: '#/components/schemas/CustomGuardrailConfig' agent_id: type: string description: Agent ID of the procedure version_id: type: - string - 'null' description: >- Version ID of a version of the procedure. None for a procedure never versioned. required: - procedure_id - name - content - agent_id title: ProcedureAtVersion-Input AgentTransfer: type: object properties: agent_id: type: - string - 'null' node_id: type: - string - 'null' condition: type: string delay_ms: type: integer default: 0 transfer_message: type: - string - 'null' enable_transferred_agent_first_message: type: boolean default: false is_workflow_node_transfer: type: boolean default: false required: - condition title: AgentTransfer PhoneNumberTransferCustomSipHeadersItems: oneOf: - type: object properties: type: type: string enum: - dynamic description: 'Discriminator value: dynamic' key: type: string description: The SIP header name (e.g., 'X-Customer-ID') value: type: string description: The dynamic variable name to resolve required: - type - key - value description: >- Custom SIP header for phone transfers with a dynamic variable reference. The value is a variable name that will be resolved at runtime. Value is not validated here since it will be substituted with actual value later. - type: object properties: type: type: string enum: - static description: 'Discriminator value: static' key: type: string description: The SIP header name (e.g., 'X-Customer-ID') value: type: string description: The header value required: - type - key - value description: >- Custom SIP header for phone transfers with a static (validated) value. discriminator: propertyName: type title: PhoneNumberTransferCustomSipHeadersItems PhoneNumberTransferTransferDestination: oneOf: - type: object properties: type: type: string enum: - phone description: 'Discriminator value: phone' phone_number: type: string required: - type - phone_number description: PhoneNumberTransferDestination variant - type: object properties: type: type: string enum: - phone_dynamic_variable description: 'Discriminator value: phone_dynamic_variable' phone_number: type: string required: - type - phone_number description: PhoneNumberDynamicVariableTransferDestination variant - type: object properties: type: type: string enum: - sip_uri description: 'Discriminator value: sip_uri' sip_uri: type: string required: - type - sip_uri description: SIPUriTransferDestination variant - type: object properties: type: type: string enum: - sip_uri_dynamic_variable description: 'Discriminator value: sip_uri_dynamic_variable' sip_uri: type: string required: - type - sip_uri description: SIPUriDynamicVariableTransferDestination variant discriminator: propertyName: type title: PhoneNumberTransferTransferDestination TransferTypeEnum: type: string enum: - blind - conference - sip_refer default: conference title: TransferTypeEnum PhoneNumberTransferPostDialDigits: oneOf: - type: object properties: type: type: string enum: - dynamic description: 'Discriminator value: dynamic' value: type: string description: The dynamic variable name to resolve required: - type - value description: PostDialDigitsDynamicVariable variant - type: object properties: type: type: string enum: - static description: 'Discriminator value: static' value: type: string description: >- DTMF digits to send after call connects (e.g., 'ww1234' for extension) required: - type - value description: PostDialDigitsStatic variant discriminator: propertyName: type description: >- DTMF digits to send after call connects (e.g., 'ww1234' for extension). Can be either a static value or a dynamic variable reference. Use 'w' for 0.5s pause. Only supported for Twilio transfers. title: PhoneNumberTransferPostDialDigits PhoneNumberTransfer: type: object properties: custom_sip_headers: type: array items: $ref: '#/components/schemas/PhoneNumberTransferCustomSipHeadersItems' description: >- Custom SIP headers to include when transferring the call. Each header can be either a static value or a dynamic variable reference. transfer_destination: $ref: '#/components/schemas/PhoneNumberTransferTransferDestination' transfer_type: $ref: '#/components/schemas/TransferTypeEnum' default: conference post_dial_digits: oneOf: - $ref: '#/components/schemas/PhoneNumberTransferPostDialDigits' - type: 'null' description: >- DTMF digits to send after call connects (e.g., 'ww1234' for extension). Can be either a static value or a dynamic variable reference. Use 'w' for 0.5s pause. Only supported for Twilio transfers. phone_number: type: - string - 'null' condition: type: string required: - transfer_destination - condition title: PhoneNumberTransfer ToolRequestModelToolConfigDiscriminatorMappingSystemParams: oneOf: - type: object properties: system_tool_type: type: string enum: - end_call description: 'Discriminator value: end_call' required: - system_tool_type description: EndCallToolConfig variant - type: object properties: system_tool_type: type: string enum: - knowledge_base_rag description: 'Discriminator value: knowledge_base_rag' required: - system_tool_type description: KnowledgeBaseRagToolConfig variant - type: object properties: system_tool_type: type: string enum: - language_detection description: 'Discriminator value: language_detection' required: - system_tool_type description: LanguageDetectionToolConfig variant - type: object properties: system_tool_type: type: string enum: - load_procedure description: 'Discriminator value: load_procedure' procedures: type: object additionalProperties: $ref: '#/components/schemas/ProcedureAtVersion-Input' required: - system_tool_type description: LoadProcedureToolConfig variant - type: object properties: system_tool_type: type: string enum: - play_keypad_touch_tone description: 'Discriminator value: play_keypad_touch_tone' use_out_of_band_dtmf: type: boolean default: true description: >- Send DTMF tones as out-of-band RTP events (RFC 4733) instead of in-band audio. Only effective for SIP trunk imported numbers. suppress_turn_after_dtmf: type: boolean default: false description: >- If true, the agent will not generate further speech after playing DTMF tones. This prevents the agent's speech from interfering with IVR systems. required: - system_tool_type description: >- Allows the agent to play DTMF tones during a phone call. This tool can be used to interact with automated phone systems, such as navigating phone menus, entering extensions, or inputting numeric codes. - type: object properties: system_tool_type: type: string enum: - skip_turn description: 'Discriminator value: skip_turn' required: - system_tool_type description: >- Allows the agent to explicitly skip its turn. This tool should be invoked by the LLM when the user indicates they would like to think or take a short pause before continuing the conversation—e.g. when they say: "Give me a second", "Let me think", or "One moment please". After calling this tool, the assistant should not speak until the user speaks again, or another normal turn-taking condition is met. The tool itself has no parameters and performs no side-effects other than informing the backend that the current turn generation is complete. - type: object properties: system_tool_type: type: string enum: - transfer_to_agent default: transfer_to_agent transfers: type: array items: $ref: '#/components/schemas/AgentTransfer' required: - system_tool_type - transfers description: TransferToAgentToolConfig variant - type: object properties: system_tool_type: type: string enum: - transfer_to_number default: transfer_to_number transfers: type: array items: $ref: '#/components/schemas/PhoneNumberTransfer' enable_client_message: type: boolean default: true description: >- Whether to play a message to the client while they wait for transfer. Defaults to true for backward compatibility. required: - system_tool_type - transfers description: TransferToNumberToolConfig variant - type: object properties: system_tool_type: type: string enum: - voicemail_detection default: voicemail_detection voicemail_message: type: - string - 'null' description: >- Optional message to leave on voicemail when detected. If not provided, the call will end immediately when voicemail is detected. Supports dynamic variables (e.g., {{system__time}}, {{system__call_duration_secs}}, {{custom_variable}}). required: - system_tool_type description: >- Allows the agent to detect when a voicemail system is encountered. This tool should be invoked by the LLM when it detects that the call has been answered by a voicemail system rather than a human. If a voicemail message is configured, it will be played; otherwise the call will end immediately. discriminator: propertyName: system_tool_type title: ToolRequestModelToolConfigDiscriminatorMappingSystemParams ConvAISecretLocator: type: object properties: secret_id: type: string required: - secret_id description: Used to reference a secret from the agent's secret store. title: ConvAISecretLocator ConvAIDynamicVariable: type: object properties: variable_name: type: string required: - variable_name description: Used to reference a dynamic variable. title: ConvAIDynamicVariable ConvAIEnvVarLocator: type: object properties: env_var_label: type: string required: - env_var_label description: Used to reference an environment variable by label. title: ConvAIEnvVarLocator WebhookToolApiSchemaConfigInputRequestHeaders: oneOf: - type: string - $ref: '#/components/schemas/ConvAISecretLocator' - $ref: '#/components/schemas/ConvAIDynamicVariable' - $ref: '#/components/schemas/ConvAIEnvVarLocator' title: WebhookToolApiSchemaConfigInputRequestHeaders WebhookToolApiSchemaConfigInputMethod: type: string enum: - GET - POST - PUT - PATCH - DELETE default: GET description: The HTTP method to use for the webhook title: WebhookToolApiSchemaConfigInputMethod QueryParamsJsonSchema: type: object properties: properties: type: object additionalProperties: $ref: '#/components/schemas/LiteralJsonSchemaProperty' required: type: array items: type: string required: - properties title: QueryParamsJsonSchema WebhookToolApiSchemaConfigInputContentType: type: string enum: - application/json - application/x-www-form-urlencoded default: application/json description: >- Content type for the request body. Only applies to POST/PUT/PATCH requests. title: WebhookToolApiSchemaConfigInputContentType AuthConnectionLocator: type: object properties: auth_connection_id: type: string required: - auth_connection_id description: >- Used to reference an auth connection from the workspace's auth connection store. title: AuthConnectionLocator EnvironmentAuthConnectionLocator: type: object properties: env_var_label: type: string required: - env_var_label description: |- References an environment variable of type 'auth_connection' by label. At runtime, resolves to the auth connection for the current environment, falling back to the default environment. title: EnvironmentAuthConnectionLocator WebhookToolApiSchemaConfigInputAuthConnection: oneOf: - $ref: '#/components/schemas/AuthConnectionLocator' - $ref: '#/components/schemas/EnvironmentAuthConnectionLocator' description: Optional auth connection to use for authentication with this webhook title: WebhookToolApiSchemaConfigInputAuthConnection WebhookToolApiSchemaConfig-Input: type: object properties: request_headers: type: object additionalProperties: $ref: '#/components/schemas/WebhookToolApiSchemaConfigInputRequestHeaders' description: Headers that should be included in the request url: type: string description: >- The URL that the webhook will be sent to. May include path parameters, e.g. https://example.com/agents/{agent_id} method: $ref: '#/components/schemas/WebhookToolApiSchemaConfigInputMethod' default: GET description: The HTTP method to use for the webhook path_params_schema: type: object additionalProperties: $ref: '#/components/schemas/LiteralJsonSchemaProperty' description: >- Schema for path parameters, if any. The keys should match the placeholders in the URL. query_params_schema: oneOf: - $ref: '#/components/schemas/QueryParamsJsonSchema' - type: 'null' description: >- Schema for any query params, if any. These will be added to end of the URL as query params. Note: properties in a query param must all be literal types request_body_schema: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Input' - type: 'null' description: >- Schema for the body parameters, if any. Used for POST/PATCH/PUT requests. The schema should be an object which will be sent as the json body response_body_schema: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Input' - type: 'null' description: >- Schema describing the expected response body structure. For documentation only; not surfaced to the LLM. content_type: $ref: '#/components/schemas/WebhookToolApiSchemaConfigInputContentType' default: application/json description: >- Content type for the request body. Only applies to POST/PUT/PATCH requests. auth_resolved_params: type: array items: type: string description: >- URL placeholders resolved from the auth connection (e.g. secrets injected via UrlSecretAuthConnection) rather than from path_params_schema. auth_connection: oneOf: - $ref: >- #/components/schemas/WebhookToolApiSchemaConfigInputAuthConnection - type: 'null' description: Optional auth connection to use for authentication with this webhook required: - url title: WebhookToolApiSchemaConfig-Input ToolRequestModelToolConfig: oneOf: - type: object properties: type: type: string enum: - client description: 'Discriminator value: client' name: type: string description: type: string description: Description of when the tool should be used and what it does. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. Must be between 1 and 120 seconds (inclusive). disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. parameters: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Input' - type: 'null' description: Schema for any parameters to pass to the client expects_response: type: boolean default: false description: >- If true, calling this tool should block the conversation until the client responds with some response which is passed to the llm. If false then we will continue the conversation without waiting for the client to respond, this is useful to show content to a user but not block the conversation dynamic_variables: $ref: '#/components/schemas/DynamicVariablesConfig-Input' description: Configuration for dynamic variables execution_mode: $ref: '#/components/schemas/ToolExecutionMode' default: immediate description: >- Determines when and how the tool executes: 'immediate' executes the tool right away when requested by the LLM, 'post_tool_speech' waits for the agent to finish speaking before executing, 'async' runs the tool in the background without blocking - best for long-running operations. required: - type - name - description description: >- A client tool is one that sends an event to the user's client to trigger something client side - type: object properties: type: type: string enum: - mcp description: 'Discriminator value: mcp' required: - type description: mcp variant - type: object properties: type: type: string enum: - system description: 'Discriminator value: system' name: type: string description: type: string default: '' description: >- Description of when the tool should be used and what it does. Leave empty to use the default description that's optimized for the specific tool type. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. params: $ref: >- #/components/schemas/ToolRequestModelToolConfigDiscriminatorMappingSystemParams required: - type - name - params description: >- A system tool is a tool that is used to call a system method in the server - type: object properties: type: type: string enum: - webhook default: webhook description: The type of tool name: type: string description: type: string description: Description of when the tool should be used and what it does. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. Must be between 5 and 120 seconds (inclusive). disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. dynamic_variables: $ref: '#/components/schemas/DynamicVariablesConfig-Input' description: Configuration for dynamic variables execution_mode: $ref: '#/components/schemas/ToolExecutionMode' default: immediate description: >- Determines when and how the tool executes: 'immediate' executes the tool right away when requested by the LLM, 'post_tool_speech' waits for the agent to finish speaking before executing, 'async' runs the tool in the background without blocking - best for long-running operations. api_schema: $ref: '#/components/schemas/WebhookToolApiSchemaConfig-Input' description: >- The schema for the outgoing webhoook, including parameters and URL specification required: - type - name - description - api_schema description: >- A webhook tool is a tool that calls an external webhook from our server discriminator: propertyName: type description: Configuration for the tool title: ToolRequestModelToolConfig UnitTestToolCallParameterEval: oneOf: - type: object properties: type: type: string enum: - anything description: 'Discriminator value: anything' required: - type description: MatchAnythingParameterEvaluationStrategy variant - type: object properties: type: type: string enum: - exact description: 'Discriminator value: exact' expected_value: type: string description: The exact string value that the parameter must match. required: - type - expected_value description: ExactParameterEvaluationStrategy variant - type: object properties: type: type: string enum: - llm description: 'Discriminator value: llm' description: type: string description: A description of the evaluation strategy to use for the test. required: - type - description description: LLMParameterEvaluationStrategy variant - type: object properties: type: type: string enum: - regex description: 'Discriminator value: regex' pattern: type: string description: A regex pattern to match the agent's response against. required: - type - pattern description: RegexParameterEvaluationStrategy variant discriminator: propertyName: type title: UnitTestToolCallParameterEval UnitTestToolCallParameter: type: object properties: eval: $ref: '#/components/schemas/UnitTestToolCallParameterEval' path: type: string required: - eval - path title: UnitTestToolCallParameter ToolResponseMockConfig-Input: type: object properties: parameter_conditions: type: array items: $ref: '#/components/schemas/UnitTestToolCallParameter' description: If the list is empty, the mock will always activate. mock_result: type: string description: The return value the LLM sees when this mock is active. required: - mock_result title: ToolResponseMockConfig-Input ToolRequestModel: type: object properties: tool_config: $ref: '#/components/schemas/ToolRequestModelToolConfig' description: Configuration for the tool response_mocks: type: - array - 'null' items: $ref: '#/components/schemas/ToolResponseMockConfig-Input' description: >- Mock responses with optional parameter conditions. Evaluated top-to-bottom; first match wins. required: - tool_config title: ToolRequestModel ArrayJsonSchemaPropertyOutputItems: oneOf: - $ref: '#/components/schemas/LiteralJsonSchemaProperty' - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Output' - $ref: '#/components/schemas/ArrayJsonSchemaProperty-Output' description: Schema for array elements. title: ArrayJsonSchemaPropertyOutputItems ArrayJsonSchemaPropertyOutputConstantValueItems: oneOf: - type: string - type: integer - type: number format: double - type: boolean title: ArrayJsonSchemaPropertyOutputConstantValueItems ArrayJsonSchemaProperty-Output: type: object properties: type: type: string enum: - array default: array description: type: string default: '' items: $ref: '#/components/schemas/ArrayJsonSchemaPropertyOutputItems' default: type: string description: Array element is_system_provided: false dynamic_variable: '' constant_value: '' is_omitted: false description: Schema for array elements. dynamic_variable: type: string default: '' description: >- When set, the entire array is populated from this dynamic variable at runtime. Mutually exclusive with description (LLM-provided array) and constant_value. constant_value: type: - array - 'null' items: $ref: >- #/components/schemas/ArrayJsonSchemaPropertyOutputConstantValueItems description: >- When set, the entire array uses this constant value at runtime. Mutually exclusive with description (LLM-provided array) and dynamic_variable. title: ArrayJsonSchemaProperty-Output ObjectJsonSchemaPropertyOutput: oneOf: - $ref: '#/components/schemas/LiteralJsonSchemaProperty' - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Output' - $ref: '#/components/schemas/ArrayJsonSchemaProperty-Output' title: ObjectJsonSchemaPropertyOutput ObjectJsonSchemaProperty-Output: type: object properties: type: type: string enum: - object default: object required: type: array items: type: string description: type: string default: '' properties: type: object additionalProperties: $ref: '#/components/schemas/ObjectJsonSchemaPropertyOutput' required_constraints: oneOf: - $ref: '#/components/schemas/RequiredConstraints' - type: 'null' title: ObjectJsonSchemaProperty-Output DynamicVariableNestedValueType-Output: description: Any type title: DynamicVariableNestedValueType-Output DynamicVariableValueTypeOutput4: type: array items: $ref: '#/components/schemas/DynamicVariableNestedValueType-Output' title: DynamicVariableValueTypeOutput4 DynamicVariableValueTypeOutput: oneOf: - type: string - type: number format: double - type: integer - type: boolean - $ref: '#/components/schemas/DynamicVariableValueTypeOutput4' title: DynamicVariableValueTypeOutput DynamicVariableValueType-Output: oneOf: - $ref: '#/components/schemas/DynamicVariableValueTypeOutput' - type: 'null' title: DynamicVariableValueType-Output DynamicVariablesConfig-Output: type: object properties: dynamic_variable_placeholders: type: object additionalProperties: $ref: '#/components/schemas/DynamicVariableValueType-Output' description: A dictionary of dynamic variable placeholders and their values title: DynamicVariablesConfig-Output ProcedureAtVersion-Output: type: object properties: procedure_id: type: string description: Procedure ID name: type: string description: Procedure name content: type: string description: Procedure content type: $ref: '#/components/schemas/ProcedureType' default: free_form guardrails: type: array items: $ref: '#/components/schemas/CustomGuardrailConfig' agent_id: type: string description: Agent ID of the procedure version_id: type: - string - 'null' description: >- Version ID of a version of the procedure. None for a procedure never versioned. required: - procedure_id - name - content - agent_id title: ProcedureAtVersion-Output ToolResponseModelToolConfigDiscriminatorMappingSystemParams: oneOf: - type: object properties: system_tool_type: type: string enum: - end_call description: 'Discriminator value: end_call' required: - system_tool_type description: EndCallToolConfig variant - type: object properties: system_tool_type: type: string enum: - knowledge_base_rag description: 'Discriminator value: knowledge_base_rag' required: - system_tool_type description: KnowledgeBaseRagToolConfig variant - type: object properties: system_tool_type: type: string enum: - language_detection description: 'Discriminator value: language_detection' required: - system_tool_type description: LanguageDetectionToolConfig variant - type: object properties: system_tool_type: type: string enum: - load_procedure description: 'Discriminator value: load_procedure' procedures: type: object additionalProperties: $ref: '#/components/schemas/ProcedureAtVersion-Output' required: - system_tool_type description: LoadProcedureToolConfig variant - type: object properties: system_tool_type: type: string enum: - play_keypad_touch_tone description: 'Discriminator value: play_keypad_touch_tone' use_out_of_band_dtmf: type: boolean default: true description: >- Send DTMF tones as out-of-band RTP events (RFC 4733) instead of in-band audio. Only effective for SIP trunk imported numbers. suppress_turn_after_dtmf: type: boolean default: false description: >- If true, the agent will not generate further speech after playing DTMF tones. This prevents the agent's speech from interfering with IVR systems. required: - system_tool_type description: >- Allows the agent to play DTMF tones during a phone call. This tool can be used to interact with automated phone systems, such as navigating phone menus, entering extensions, or inputting numeric codes. - type: object properties: system_tool_type: type: string enum: - skip_turn description: 'Discriminator value: skip_turn' required: - system_tool_type description: >- Allows the agent to explicitly skip its turn. This tool should be invoked by the LLM when the user indicates they would like to think or take a short pause before continuing the conversation—e.g. when they say: "Give me a second", "Let me think", or "One moment please". After calling this tool, the assistant should not speak until the user speaks again, or another normal turn-taking condition is met. The tool itself has no parameters and performs no side-effects other than informing the backend that the current turn generation is complete. - type: object properties: system_tool_type: type: string enum: - transfer_to_agent default: transfer_to_agent transfers: type: array items: $ref: '#/components/schemas/AgentTransfer' required: - system_tool_type - transfers description: TransferToAgentToolConfig variant - type: object properties: system_tool_type: type: string enum: - transfer_to_number default: transfer_to_number transfers: type: array items: $ref: '#/components/schemas/PhoneNumberTransfer' enable_client_message: type: boolean default: true description: >- Whether to play a message to the client while they wait for transfer. Defaults to true for backward compatibility. required: - system_tool_type - transfers description: TransferToNumberToolConfig variant - type: object properties: system_tool_type: type: string enum: - voicemail_detection default: voicemail_detection voicemail_message: type: - string - 'null' description: >- Optional message to leave on voicemail when detected. If not provided, the call will end immediately when voicemail is detected. Supports dynamic variables (e.g., {{system__time}}, {{system__call_duration_secs}}, {{custom_variable}}). required: - system_tool_type description: >- Allows the agent to detect when a voicemail system is encountered. This tool should be invoked by the LLM when it detects that the call has been answered by a voicemail system rather than a human. If a voicemail message is configured, it will be played; otherwise the call will end immediately. discriminator: propertyName: system_tool_type title: ToolResponseModelToolConfigDiscriminatorMappingSystemParams WebhookToolApiSchemaConfigOutputRequestHeaders: oneOf: - type: string - $ref: '#/components/schemas/ConvAISecretLocator' - $ref: '#/components/schemas/ConvAIDynamicVariable' - $ref: '#/components/schemas/ConvAIEnvVarLocator' title: WebhookToolApiSchemaConfigOutputRequestHeaders WebhookToolApiSchemaConfigOutputMethod: type: string enum: - GET - POST - PUT - PATCH - DELETE default: GET description: The HTTP method to use for the webhook title: WebhookToolApiSchemaConfigOutputMethod WebhookToolApiSchemaConfigOutputContentType: type: string enum: - application/json - application/x-www-form-urlencoded default: application/json description: >- Content type for the request body. Only applies to POST/PUT/PATCH requests. title: WebhookToolApiSchemaConfigOutputContentType WebhookToolApiSchemaConfigOutputAuthConnection: oneOf: - $ref: '#/components/schemas/AuthConnectionLocator' - $ref: '#/components/schemas/EnvironmentAuthConnectionLocator' description: Optional auth connection to use for authentication with this webhook title: WebhookToolApiSchemaConfigOutputAuthConnection WebhookToolApiSchemaConfig-Output: type: object properties: request_headers: type: object additionalProperties: $ref: >- #/components/schemas/WebhookToolApiSchemaConfigOutputRequestHeaders description: Headers that should be included in the request url: type: string description: >- The URL that the webhook will be sent to. May include path parameters, e.g. https://example.com/agents/{agent_id} method: $ref: '#/components/schemas/WebhookToolApiSchemaConfigOutputMethod' default: GET description: The HTTP method to use for the webhook path_params_schema: type: object additionalProperties: $ref: '#/components/schemas/LiteralJsonSchemaProperty' description: >- Schema for path parameters, if any. The keys should match the placeholders in the URL. query_params_schema: oneOf: - $ref: '#/components/schemas/QueryParamsJsonSchema' - type: 'null' description: >- Schema for any query params, if any. These will be added to end of the URL as query params. Note: properties in a query param must all be literal types request_body_schema: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Output' - type: 'null' description: >- Schema for the body parameters, if any. Used for POST/PATCH/PUT requests. The schema should be an object which will be sent as the json body response_body_schema: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Output' - type: 'null' description: >- Schema describing the expected response body structure. For documentation only; not surfaced to the LLM. content_type: $ref: '#/components/schemas/WebhookToolApiSchemaConfigOutputContentType' default: application/json description: >- Content type for the request body. Only applies to POST/PUT/PATCH requests. auth_resolved_params: type: array items: type: string description: >- URL placeholders resolved from the auth connection (e.g. secrets injected via UrlSecretAuthConnection) rather than from path_params_schema. auth_connection: oneOf: - $ref: >- #/components/schemas/WebhookToolApiSchemaConfigOutputAuthConnection - type: 'null' description: Optional auth connection to use for authentication with this webhook required: - url title: WebhookToolApiSchemaConfig-Output ToolResponseModelToolConfig: oneOf: - type: object properties: type: type: string enum: - client description: 'Discriminator value: client' name: type: string description: type: string description: Description of when the tool should be used and what it does. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. Must be between 1 and 120 seconds (inclusive). disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. parameters: oneOf: - $ref: '#/components/schemas/ObjectJsonSchemaProperty-Output' - type: 'null' description: Schema for any parameters to pass to the client expects_response: type: boolean default: false description: >- If true, calling this tool should block the conversation until the client responds with some response which is passed to the llm. If false then we will continue the conversation without waiting for the client to respond, this is useful to show content to a user but not block the conversation dynamic_variables: $ref: '#/components/schemas/DynamicVariablesConfig-Output' description: Configuration for dynamic variables execution_mode: $ref: '#/components/schemas/ToolExecutionMode' default: immediate description: >- Determines when and how the tool executes: 'immediate' executes the tool right away when requested by the LLM, 'post_tool_speech' waits for the agent to finish speaking before executing, 'async' runs the tool in the background without blocking - best for long-running operations. required: - type - name - description description: >- A client tool is one that sends an event to the user's client to trigger something client side - type: object properties: type: type: string enum: - mcp description: 'Discriminator value: mcp' required: - type description: mcp variant - type: object properties: type: type: string enum: - system description: 'Discriminator value: system' name: type: string description: type: string default: '' description: >- Description of when the tool should be used and what it does. Leave empty to use the default description that's optimized for the specific tool type. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. params: $ref: >- #/components/schemas/ToolResponseModelToolConfigDiscriminatorMappingSystemParams required: - type - name - params description: >- A system tool is a tool that is used to call a system method in the server - type: object properties: type: type: string enum: - webhook default: webhook description: The type of tool name: type: string description: type: string description: Description of when the tool should be used and what it does. response_timeout_secs: type: integer default: 20 description: >- The maximum time in seconds to wait for the tool call to complete. Must be between 5 and 120 seconds (inclusive). disable_interruptions: type: boolean default: false description: >- If true, the user will not be able to interrupt the agent while this tool is running. force_pre_tool_speech: type: boolean default: false description: >- DEPRECATED: use `pre_tool_speech` instead. If true, the agent will speak before the tool call. pre_tool_speech: $ref: '#/components/schemas/PreToolSpeechMode' default: auto description: >- Controls whether the agent speaks before this tool is called. 'auto' (default) decides based on recent tool latency, 'force' always asks the agent to speak, 'off' fully opts out regardless of latency. assignments: type: array items: $ref: '#/components/schemas/DynamicVariableAssignment' description: >- Configuration for extracting values from tool responses and assigning them to dynamic variables tool_call_sound: oneOf: - $ref: '#/components/schemas/ToolCallSoundType' - type: 'null' description: >- Predefined tool call sound type to play during tool execution. If not specified, no tool call sound will be played. tool_call_sound_behavior: $ref: '#/components/schemas/ToolCallSoundBehavior' default: auto description: >- Determines when the tool call sound should play. 'auto' only plays when there's pre-tool speech, 'always' plays for every tool call. tool_error_handling_mode: $ref: '#/components/schemas/ToolErrorHandlingMode' default: auto description: >- Controls how tool errors are processed before being shared with the agent. 'auto' determines handling based on tool type (summarized for native integrations, hide for others), 'summarized' sends an LLM-generated summary, 'passthrough' sends the raw error, 'hide' does not share the error with the agent. dynamic_variables: $ref: '#/components/schemas/DynamicVariablesConfig-Output' description: Configuration for dynamic variables execution_mode: $ref: '#/components/schemas/ToolExecutionMode' default: immediate description: >- Determines when and how the tool executes: 'immediate' executes the tool right away when requested by the LLM, 'post_tool_speech' waits for the agent to finish speaking before executing, 'async' runs the tool in the background without blocking - best for long-running operations. api_schema: $ref: '#/components/schemas/WebhookToolApiSchemaConfig-Output' description: >- The schema for the outgoing webhoook, including parameters and URL specification required: - type - name - description - api_schema description: >- A webhook tool is a tool that calls an external webhook from our server discriminator: propertyName: type description: The type of tool title: ToolResponseModelToolConfig ResourceAccessInfoRole: type: string enum: - admin - editor - commenter - viewer description: The role of the user making the request title: ResourceAccessInfoRole ResourceAccessInfoAnonymousAccessLevelOverride: type: string enum: - admin - editor - commenter - viewer description: >- The access level for anonymous users. If None, the resource is not shared publicly. title: ResourceAccessInfoAnonymousAccessLevelOverride ResourceAccessInfoAccessSource: type: string enum: - creator - explicit - workspace_admin - workspace_default description: >- Why the requesting user has access to this resource. 'creator' = caller is the owner. 'explicit' = caller (or one of their workspace groups) is listed in role_to_group_ids beyond the workspace-wide everyone group. 'workspace_default' = the workspace-wide everyone group is listed in role_to_group_ids (every non-anon workspace member, including admins, sees this resource). 'workspace_admin' = caller is a workspace admin and the admin seat is the *only* path to access; reserved for docs nobody else can see. Lets the UI disclose why an admin-bypass viewer sees a doc that wasn't explicitly shared with them. title: ResourceAccessInfoAccessSource ResourceAccessInfo: type: object properties: is_creator: type: boolean description: Whether the user making the request is the creator of the agent creator_name: type: string description: Name of the agent's creator creator_email: type: string description: Email of the agent's creator role: $ref: '#/components/schemas/ResourceAccessInfoRole' description: The role of the user making the request anonymous_access_level_override: oneOf: - $ref: >- #/components/schemas/ResourceAccessInfoAnonymousAccessLevelOverride - type: 'null' description: >- The access level for anonymous users. If None, the resource is not shared publicly. access_source: oneOf: - $ref: '#/components/schemas/ResourceAccessInfoAccessSource' - type: 'null' description: >- Why the requesting user has access to this resource. 'creator' = caller is the owner. 'explicit' = caller (or one of their workspace groups) is listed in role_to_group_ids beyond the workspace-wide everyone group. 'workspace_default' = the workspace-wide everyone group is listed in role_to_group_ids (every non-anon workspace member, including admins, sees this resource). 'workspace_admin' = caller is a workspace admin and the admin seat is the *only* path to access; reserved for docs nobody else can see. Lets the UI disclose why an admin-bypass viewer sees a doc that wasn't explicitly shared with them. required: - is_creator - creator_name - creator_email - role title: ResourceAccessInfo ToolUsageStatsResponseModel: type: object properties: total_calls: type: integer default: 0 description: The total number of calls to the tool avg_latency_secs: type: number format: double required: - avg_latency_secs title: ToolUsageStatsResponseModel ToolResponseMockConfig-Output: type: object properties: parameter_conditions: type: array items: $ref: '#/components/schemas/UnitTestToolCallParameter' description: If the list is empty, the mock will always activate. mock_result: type: string description: The return value the LLM sees when this mock is active. required: - mock_result title: ToolResponseMockConfig-Output ToolResponseModel: type: object properties: id: type: string tool_config: $ref: '#/components/schemas/ToolResponseModelToolConfig' description: The type of tool access_info: $ref: '#/components/schemas/ResourceAccessInfo' usage_stats: $ref: '#/components/schemas/ToolUsageStatsResponseModel' response_mocks: type: - array - 'null' items: $ref: '#/components/schemas/ToolResponseMockConfig-Output' description: >- Mock responses with optional parameter conditions. Evaluated top-to-bottom; first match wins. required: - id - tool_config - access_info - usage_stats title: ToolResponseModel ValidationErrorLocItems: oneOf: - type: string - type: integer title: ValidationErrorLocItems ValidationError: type: object properties: loc: type: array items: $ref: '#/components/schemas/ValidationErrorLocItems' msg: type: string type: type: string required: - loc - msg - type title: ValidationError HTTPValidationError: type: object properties: detail: type: array items: $ref: '#/components/schemas/ValidationError' title: HTTPValidationError ``` ## SDK Code Examples ```typescript import { ElevenLabsClient } from "@elevenlabs/elevenlabs-js"; async function main() { const client = new ElevenLabsClient(); await client.conversationalAi.tools.create({ toolConfig: { type: "system", name: "end_call", params: { systemToolType: "end_call", }, description: "", }, }); } main(); ``` ```python from elevenlabs import ElevenLabs, ToolRequestModel, ToolRequestModelToolConfig_System, SystemToolConfigInputParams_EndCall client = ElevenLabs() client.conversational_ai.tools.create( request=ToolRequestModel( tool_config=ToolRequestModelToolConfig_System( name="end_call", params=SystemToolConfigInputParams_EndCall(), description="", ), ), ) ``` ```go package main import ( "fmt" "strings" "net/http" "io" ) func main() { url := "https://api.elevenlabs.io/v1/convai/tools" payload := strings.NewReader("{\n \"tool_config\": {\n \"type\": \"system\",\n \"name\": \"end_call\",\n \"params\": {\n \"system_tool_type\": \"end_call\"\n },\n \"description\": \"\"\n }\n}") req, _ := http.NewRequest("POST", url, payload) req.Header.Add("Content-Type", "application/json") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("https://api.elevenlabs.io/v1/convai/tools") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Post.new(url) request["Content-Type"] = 'application/json' request.body = "{\n \"tool_config\": {\n \"type\": \"system\",\n \"name\": \"end_call\",\n \"params\": {\n \"system_tool_type\": \"end_call\"\n },\n \"description\": \"\"\n }\n}" response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.post("https://api.elevenlabs.io/v1/convai/tools") .header("Content-Type", "application/json") .body("{\n \"tool_config\": {\n \"type\": \"system\",\n \"name\": \"end_call\",\n \"params\": {\n \"system_tool_type\": \"end_call\"\n },\n \"description\": \"\"\n }\n}") .asString(); ``` ```php request('POST', 'https://api.elevenlabs.io/v1/convai/tools', [ 'body' => '{ "tool_config": { "type": "system", "name": "end_call", "params": { "system_tool_type": "end_call" }, "description": "" } }', 'headers' => [ 'Content-Type' => 'application/json', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://api.elevenlabs.io/v1/convai/tools"); var request = new RestRequest(Method.POST); request.AddHeader("Content-Type", "application/json"); request.AddParameter("application/json", "{\n \"tool_config\": {\n \"type\": \"system\",\n \"name\": \"end_call\",\n \"params\": {\n \"system_tool_type\": \"end_call\"\n },\n \"description\": \"\"\n }\n}", ParameterType.RequestBody); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = ["Content-Type": "application/json"] let parameters = ["tool_config": [ "type": "system", "name": "end_call", "params": ["system_tool_type": "end_call"], "description": "" ]] as [String : Any] let postData = JSONSerialization.data(withJSONObject: parameters, options: []) let request = NSMutableURLRequest(url: NSURL(string: "https://api.elevenlabs.io/v1/convai/tools")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "POST" request.allHTTPHeaderFields = headers request.httpBody = postData as Data let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ```