marcela-ai

Configure a multi-channel AI assistant with ElevenLabs, Twilio, and RAG.

Updated Mar 10, 2026
One-click install
npx skills add https://github.com/ricardocidale/hbg-screens --skill marcela-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: marcela-ai
Source: https://github.com/ricardocidale/hbg-screens/tree/main/.claude/skills/marcela-ai
Command: npx skills add https://github.com/ricardocidale/hbg-screens --skill marcela-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a comprehensive, multi-channel AI conversational assistant that can interact with users via web chat, phone calls, and SMS, offering a unified and intelligent user experience.

Core Features & Use Cases

  • Multi-Channel Support: Seamlessly operates across web widgets, Twilio Voice, and Twilio SMS.
  • Configurable AI Agent: Allows customization of the AI's name, voice, LLM, and system prompts.
  • Integrated Knowledge Base: Leverages Retrieval-Augmented Generation (RAG) for informed responses.
  • Deterministic Tools: Provides access to financial data and other tools with precise, reliable outputs.
  • Use Case: Deploy an AI assistant named "Concierge" that can answer user queries via a web chat widget, handle customer service calls, and send SMS notifications, all while drawing from a curated knowledge base and providing accurate financial insights.

Quick Start

Configure the AI agent's name and voice settings through the admin interface.

Frequently Asked Questions about marcela-ai

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a multi-channel voice assistant that handles web chat, SMS, and phone calls?

A multi-channel voice assistant integrates ElevenLabs for voice generation, Twilio for telephony and SMS, and a RAG knowledge base to unify web widget, phone, and text interactions under a configurable AI agent.

Can I use Twilio and ElevenLabs together for a conversational AI agent?

Yes, Twilio and ElevenLabs can be combined for conversational AI by using Twilio for telephony and SMS routing and ElevenLabs for voice synthesis, managed through a centralized server-side audio pipeline and system prompts.

How does a RAG knowledge base work with an AI chatbot for customer service?

A RAG knowledge base supplies curated context to an AI chatbot, enabling the conversational agent to retrieve specific information and provide informed responses across web chat and SMS channels.

What is the best way to configure system prompts and voices for an AI agent?

Configuring system prompts and voices for an AI agent is best handled through a centralized admin interface, allowing you to customize the assistant's name, LLM behavior, and ElevenLabs voice settings before deployment.

Does the ElevenLabs voice assistant support deterministic tools for financial data?

Yes, the ElevenLabs voice assistant supports deterministic tools, providing precise and reliable outputs for financial data queries alongside standard conversational responses across web and phone channels.

What are the limitations of using SMS for RAG-based AI agent conversations?

Using SMS for RAG-based AI agent conversations limits interactions to text-only exchanges, lacking the voice capabilities available through Twilio Voice and web widgets, which may constrain complex audio pipeline interactions.