agents

Create and manage real-time voice AI agents using the ElevenLabs Conversational AI platform.

Updated Jul 16, 2026
One-click install
npx skills add https://github.com/X-manist/Cohmira --skill agents-x-manist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agents
Source: https://github.com/X-manist/Cohmira/tree/main/src/builtin-plugins/openmontage/.agents/skills/agents
Command: npx skills add https://github.com/X-manist/Cohmira --skill agents-x-manist

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @elevenlabs/cli, @elevenlabs/elevenlabs-js, @elevenlabs/client, @elevenlabs/react, elevenlabs, and includes references (resource) components.

What problem does it solve? Building real-time voice assistants requires coordinating speech recognition, LLM reasoning, text-to-speech, telephony, and web embedding across many APIs. This Skill consolidates the ElevenLabs Agents platform into one workflow so you can create, configure, deploy, and manage conversational voice agents without piecing together documentation. ## Core Features & Use Cases - Agent Creation & Configuration: Create agents via CLI, Python/JavaScript SDKs, or REST API with configurable LLM providers (OpenAI, Anthropic, Google, custom endpoints), voices, turn-taking behavior, and knowledge bases. - Tools & Integrations: Extend agents with webhook tools, browser-side client tools, and built-in system tools like call transfer and end_call, plus outbound phone calls through Twilio. - Web Embedding & Management: Embed a voice widget on any website, manage agents as code with the CLI push/pull workflow, and control guardrails, privacy redaction, and call limits. - Use Case: Build a customer support voice bot that answers calls, looks up orders via a webhook tool, transfers complex cases to a human phone number, and embeds on your help page as a chat widget. ## Quick Start Create an ElevenLabs voice agent named Support Bot with a greeting, a Gemini LLM prompt, and a default voice, then show me how to embed it on my website.

Frequently Asked Questions about agents

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

FAQPage Schema
How do I create a voice AI agent with ElevenLabs?

Install the ElevenLabs CLI with npm install -g @elevenlabs/cli, run elevenlabs agents init, then elevenlabs agents add "My Assistant" --template complete, and push with elevenlabs agents push. You can also create agents programmatically via the Python or JavaScript SDKs using client.conversational_ai.agents.create.

Which LLM models can ElevenLabs agents use?

Agents support OpenAI models like gpt-4o and gpt-4.1, Anthropic Claude models, Google Gemini models such as gemini-2.0-flash, ElevenLabs-hosted open models, and custom LLM endpoints via the custom_llm configuration. Use GET /v1/convai/llm/list to inspect the current model catalog.

How do I add custom tools to an ElevenLabs voice agent?

Define webhook tools for server-side API calls and client tools for browser-side actions inside conversation_config.agent.prompt.tools, and configure system tools like end_call or transfer_to_number in built_in_tools. Webhook tools specify a URL, method, and request body schema.

Can ElevenLabs agents make outbound phone calls?

Yes, outbound calls work through Twilio integration using client.conversational_ai.twilio.outbound_call with an agent_id, agent_phone_number_id, and destination number. You can override the first message and pass dynamic variables per call.

How do I embed an ElevenLabs voice agent on my website?

Add the <elevenlabs-convai agent-id="your-agent-id"> element and load the convai-widget-embed script from unpkg. Customize appearance with attributes like avatar-image-url, action-text, and variant, or use signed URLs for authenticated agents.

What API key do ElevenLabs agents require?

All agent operations require an ElevenLabs API key set as the ELEVENLABS_API_KEY environment variable or passed directly to the SDK client. Create a key in the API Keys section of your ElevenLabs account settings.