VapiAIVapiAIOfficialยท10 Agent Skills Included

skills

Build, deploy, and test Vapi voice phone agents

Builds production voice agents on Vapi: assistants, squads, tools, phone numbers, outbound calls, and campaigns. Eliminates manual API payload writing, prompt guesswork, and unsafe live mutations through verified schemas and confirmation steps. Adds prompt auditing, post-call data extraction, webhooks, and simulation testing so teams ship reliable phone agents faster.
npx skills add VapiAI/skills --all -g -y
Available:

Tells the agent which Vapi skill to route each request to, when to use the bundled documentation MCP server for gaps, and that all API calls require the VAPI_API_KEY environment variable.

All Skills in This Repository (10)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Vapi skills?โ–ผ

Run `npx skills add VapiAI/skills --all -g -y` in your terminal to install every skill globally. You can also install a single skill with `npx skills add VapiAI/skills --skill create-assistant`.

How to build a voice AI phone agent with Vapi?โ–ผ

Use the create-assistant skill to design the agent's prompt, voice, model, and tools, then create-phone-number to attach a number and create-call to test it live. The vapi-prompt-builder skill helps you write production-quality system prompts.

How to run outbound calling campaigns with Vapi?โ–ผ

The create-campaign skill handles persistent multi-contact campaigns with scheduling, concurrency limits, CSV contacts, and webhooks. It includes safety checks and confirmation steps before any live, chargeable calls are placed.

Do Vapi skills work with Claude Code and Cursor?โ–ผ

Yes. All skills follow the universal SKILL.md standard and work in Claude Code, Cursor, VS Code Copilot, Gemini CLI, and Codex. The bundled Vapi documentation MCP server auto-configures in agents that support MCP.

How to test a Vapi voice agent before launch?โ–ผ

Use the simulations skill to run realistic chat and voice test conversations against your assistant or squad with measurable pass/fail criteria. It supports tool mocking so tests never trigger real side effects.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’