Skip to main content
Glama

create_conversation

Start a conversational video interface using AI replicas and personas to enable interactive video conversations with custom context and recording options.

Instructions

Create a new conversational video interface

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
replica_idNoReplica to use for the conversation
persona_idNoPersona to use for the conversation
conversation_nameNoName for the conversation
callback_urlNoURL to receive conversation events
conversational_contextNoContext for the conversation
custom_greetingNoCustom greeting message
enable_recordingNoEnable conversation recording

Implementation Reference

  • The createConversation handler method that executes the tool logic. It makes a POST request to the Tavus API /conversations endpoint with the provided arguments and returns the response data in MCP format.
    private async createConversation(args: any) {
      const response = await this.axiosInstance.post('/conversations', args);
      return {
        content: [{
          type: 'text',
          text: JSON.stringify(response.data, null, 2),
        }],
      };
    }
  • The input schema definition for the create_conversation tool. Defines all input parameters including replica_id, persona_id, conversation_name, callback_url, conversational_context, custom_greeting, and enable_recording with their types and descriptions.
    {
      name: 'create_conversation',
      description: 'Create a new conversational video interface',
      inputSchema: {
        type: 'object',
        properties: {
          replica_id: {
            type: 'string',
            description: 'Replica to use for the conversation',
          },
          persona_id: {
            type: 'string',
            description: 'Persona to use for the conversation',
          },
          conversation_name: {
            type: 'string',
            description: 'Name for the conversation',
          },
          callback_url: {
            type: 'string',
            description: 'URL to receive conversation events',
          },
          conversational_context: {
            type: 'string',
            description: 'Context for the conversation',
          },
          custom_greeting: {
            type: 'string',
            description: 'Custom greeting message',
          },
          enable_recording: {
            type: 'boolean',
            description: 'Enable conversation recording',
          },
        },
      },
    },
  • src/index.ts:355-391 (registration)
    The tool registration for create_conversation in the ListToolsRequestSchema handler. Registers the tool name, description, and inputSchema as part of the available tools list.
    {
      name: 'create_conversation',
      description: 'Create a new conversational video interface',
      inputSchema: {
        type: 'object',
        properties: {
          replica_id: {
            type: 'string',
            description: 'Replica to use for the conversation',
          },
          persona_id: {
            type: 'string',
            description: 'Persona to use for the conversation',
          },
          conversation_name: {
            type: 'string',
            description: 'Name for the conversation',
          },
          callback_url: {
            type: 'string',
            description: 'URL to receive conversation events',
          },
          conversational_context: {
            type: 'string',
            description: 'Context for the conversation',
          },
          custom_greeting: {
            type: 'string',
            description: 'Custom greeting message',
          },
          enable_recording: {
            type: 'boolean',
            description: 'Enable conversation recording',
          },
        },
      },
    },
  • The switch case routing for the create_conversation tool in the CallToolRequestSchema handler. Routes incoming tool calls to the createConversation method with the request arguments.
    case 'create_conversation':
      return await this.createConversation(request.params.arguments);

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/rakeshdavid/Tavus-MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server