Vapi Voice Agent Management for Openclaw

A comprehensive skill for Openclaw agents to programmatically control Vapi voice assistants, handle real-time calls, and manage telephony infrastructure.

colygon
v0.1.0
Feb 7, 2026
0
2.3k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install vapi-skill

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install vapi-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Vapi Voice Agent Management?

This skill enables Openclaw agents to interface directly with Vapi, a leading voice AI platform. It provides a robust bridge to create, update, and manage sophisticated voice assistants using various model and voice providers like OpenAI, Groq, and ElevenLabs. By leveraging this integration, developers can automate the entire lifecycle of a voice interaction, from provisioning phone numbers to executing complex tool calls within a live conversation.

Built to be flexible, this addition to the library of Openclaw Skills supports both REST-based automated workflows and interactive CLI operations. Whether you are building a customer support concierge or an automated outbound notification system, this skill provides the necessary endpoints and helper scripts to integrate high-quality voice capabilities into your AI agent's toolkit.

Vapi Voice Agent Management Use Cases

  • Automating outbound call campaigns for customer concierge or appointment reminders.
  • Programmatically creating and updating voice assistants with specific personality models and voice IDs.
  • Managing phone number inventory and linking specific numbers to specialized AI agents.
  • Monitoring call logs and webhook events for real-time business intelligence and analytics.
  • Developing interactive voice response (IVR) systems using advanced function calling and tool management.

How Vapi Voice Agent Management Works

  1. Authenticate with the Vapi platform by providing a valid VAPI_API_KEY as a gateway secret or environment variable.
  2. Select the operational mode by setting VAPI_MODE to either 'api' (REST-first) or 'cli' (interactive-first).
  3. Execute commands via the provided Node.js helper script or the official Vapi CLI to list, create, or modify assistants.
  4. Initiate outbound calls or inspect incoming call metadata by sending structured requests to the Vapi base API endpoint.
  5. Leverage MCP (Model Context Protocol) documentation support to provide your IDE with deep context for voice agent development and troubleshooting.

Vapi Voice Agent Management Setup

Set the required environment variable in your shell or gateway secrets:

export VAPI_API_KEY='your_vapi_api_key_here'

To use the REST helper script included in the Openclaw Skills repository:

# List all configured assistants
node skills/vapi/bin/vapi-api.mjs assistants:list

# Create a new concierge assistant
node skills/vapi/bin/vapi-api.mjs assistants:create --name "Claw Concierge" --modelProvider openai --model gpt-4o-mini --voiceProvider 11labs --voiceId rachel

To install and log into the Vapi CLI for interactive workflows:

curl -sSL https://vapi.ai/install.sh | bash
vapi login

Vapi Voice Agent Management Data Schema & Taxonomy

The skill interacts with Vapi's REST API and organizes data according to the following structures:

Entity Description Key Attributes
Assistants Voice agent configurations Name, Model Provider, Voice ID, Transcriber
Calls Active or historical call records Call ID, To/From Number, Status, Duration
Phone Numbers Telephony resources Number, Provider, Assistant Link
Tools Function calling definitions Name, Parameters, Server URL
Webhooks Event notification settings URL, Secret, Event Types

Vapi Voice Agent Management Advanced Features

  • Dual-mode operation supporting both direct REST API calls via Node.js and interactive CLI commands.
  • MCP (Model Context Protocol) server support for enhanced IDE assistance in Cursor, Windsurf, or VSCode.
  • Multi-provider support allowing developers to mix and match LLM backends with high-fidelity voice synthesis from ElevenLabs.
  • Complex tool-calling capabilities that allow the voice assistant to trigger external APIs during a live call.
  • Comprehensive webhook management for building event-driven architectures around voice interactions within Openclaw Skills.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*