Skip to main content
Glama

create_persona

Create conversational AI personas by configuring speech, language, and perception settings for video generation and replica management.

Instructions

Create a new persona for conversational AI

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
persona_nameNoName for the persona
replica_idNoReplica to use for this persona
contextNoContextual information for the LLM
system_promptNoSystem prompt for the LLM
layersNoConfiguration layers for the persona

Implementation Reference

  • The createPersona method is the handler that executes the create_persona tool logic. It makes a POST request to /personas endpoint with the provided arguments and returns the response data as formatted JSON text.
    private async createPersona(args: any) {
      const response = await this.axiosInstance.post('/personas', args);
      return {
        content: [{
          type: 'text',
          text: JSON.stringify(response.data, null, 2),
        }],
      };
    }
  • Tool registration with name 'create_persona', description 'Create a new persona for conversational AI', and input schema defining properties: persona_name, replica_id, context, system_prompt, and layers (with nested stt, llm, tts, perception settings).
    // Replica Personas
    {
      name: 'create_persona',
      description: 'Create a new persona for conversational AI',
      inputSchema: {
        type: 'object',
        properties: {
          persona_name: {
            type: 'string',
            description: 'Name for the persona',
          },
          replica_id: {
            type: 'string',
            description: 'Replica to use for this persona',
          },
          context: {
            type: 'string',
            description: 'Contextual information for the LLM',
          },
          system_prompt: {
            type: 'string',
            description: 'System prompt for the LLM',
          },
          layers: {
            type: 'object',
            description: 'Configuration layers for the persona',
            properties: {
              stt: { type: 'object', description: 'Speech-to-text settings' },
              llm: { type: 'object', description: 'Language model settings' },
              tts: { type: 'object', description: 'Text-to-speech settings' },
              perception: { type: 'object', description: 'Perception settings (Raven-0)' },
            },
          },
        },
      },
    },
  • src/index.ts:728-729 (registration)
    Switch case statement that routes 'create_persona' tool calls to the createPersona handler method.
    case 'create_persona':
      return await this.createPersona(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