agents

Create and manage real-time voice agents with integrated TTS and ASR.

1|Updated Apr 11, 2026
One-click install
npx skills add https://github.com/shige1014-dev/backup-OpenMontage --skill agents-shige1014-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agents
Source: https://github.com/shige1014-dev/backup-OpenMontage/tree/main/.agents/skills/agents
Command: npx skills add https://github.com/shige1014-dev/backup-OpenMontage --skill agents-shige1014-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Build, configure, and deploy conversational voice agents that provide real-time TTS/ASR interactions so teams can ship voice assistants, IVR systems, and interactive characters without building voice infrastructure from scratch.

Core Features & Use Cases

  • Multi-provider LLM integration: Use OpenAI, Anthropic, Google, or custom LLMs for agent reasoning and responses.
  • Real-time voice: Configure TTS and ASR settings, low-latency models, expressive voices, and turn-taking behavior for live conversations.
  • Tools & integrations: Add webhook and client tools, outbound Twilio calls, widget embedding for web, and CI/CD push/pull workflows.
  • Use Case: Deploy a customer support agent embedded on a website that can query backend APIs via webhooks, transfer calls, and make personalized outbound reminders.

Quick Start

Create a new ElevenLabs agent named My Assistant that greets users in English, uses a concise helpful prompt, and uses the voice JBFqnCBsd6RMkjVDRZzb.

Frequently Asked Questions about agents

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

FAQPage Schema
How do I build a real-time conversational voice AI agent?

To build a real-time conversational voice AI agent, you can configure integrated TTS and ASR settings with multi-LLM provider selection. This allows you to deploy web widget embeds, telephony outbound calls, and in-app assistants without building voice infrastructure from scratch.

Can I use Twilio for outbound calls with an ElevenLabs voice agent?

Yes, you can use Twilio for outbound calls with an ElevenLabs voice agent. The system supports telephony outbound calls alongside webhook and client tools, enabling interactive voice responses and personalized reminders.

What LLM providers can I use for voice assistant reasoning and responses?

You can use OpenAI, Anthropic, Google, or custom LLMs for voice assistant reasoning and responses. This multi-provider LLM integration allows flexible configuration of your conversational AI agent's behavior.

Do I need an ElevenLabs API key to deploy a voice agent widget?

Yes, you need an ElevenLabs API key and internet access to deploy a voice agent widget. These are required to authenticate the real-time TTS and ASR interactions that power the conversational experience.

How do I configure TTS and ASR settings for a low-latency voice assistant?

You configure TTS and ASR settings for a low-latency voice assistant by selecting expressive voices and defining turn-taking behavior. These configurable conversation settings ensure smooth, live interactions for customer support or e-commerce applications.

Can a web widget voice agent query backend APIs via webhooks?

Yes, a web widget voice agent can query backend APIs via webhooks. By adding webhook and client tools to your conversational agent, it can fetch external data, transfer calls, and perform interactive tasks during real-time conversations.