---
name: al-699-put-a-voice-agent-on-your-website-that-answers-from-your
description: "Moss's Founding Agent case study describes running a voice AI agent on its own website for five months: 3,516 conversations, 9,500+ visitor interactions, ~120 hours of live voice, 3,733 questions with ~74% answered from the knowledge base without a human, 61% of conversations outside business hours, and a 4x increase in demo bookings with higher intent. The agent answers from site/docs/FAQ knowledge, labels intent (Ready to buy through Not a fit), captures leads after sustained engagement, books demos by voice against a calendar, supports live human handoff, and feeds unanswered questions back into content gaps. Use when the user wants help with website voice agent for q&a and demo booking (sales) for a small business."
---

# Put a voice agent on your website that answers from your docs, scores buying intent and books the demo while your team is offline

Moss's Founding Agent case study describes running a voice AI agent on its own website for five months: 3,516 conversations, 9,500+ visitor interactions, ~120 hours of live voice, 3,733 questions with ~74% answered from the knowledge base without a human, 61% of conversations outside business hours, and a 4x increase in demo bookings with higher intent. The agent answers from site/docs/FAQ knowledge, labels intent (Ready to buy through Not a fit), captures leads after sustained engagement, books demos by voice against a calendar, supports live human handoff, and feeds unanswered questions back into content gaps.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- Company
- Site URLs, docs, FAQ, decks
- Tone
- Calendar
- Minutes
- Your own material (text, notes or data)

## Instructions

Do the task below with the user's inputs in place of each [bracket].

Design a website voice agent for [company] like Moss Founding Agent. Knowledge sources: [site URLs, docs, FAQ, decks]. Voice and greeting: [tone]. Rules: answer only from approved knowledge; admit gaps instead of inventing; after [N] engaged questions ask for email; if intent is Ready to buy or Evaluating, offer to book a demo on [calendar]; notify a human when a call runs longer than [minutes] or the visitor asks for a person. Define intent labels, the lead fields to capture, the weekly report (top unanswered questions, topics, demo conversions), and a 15-minute go-live checklist. Do not claim metrics we have not measured.

## Safety checks (share these with the user)

- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: Moss Founding Agent, voice AI, calendar, website knowledge base
- Based on this real-world example: https://www.moss.dev/blog/founding-agent-case-study
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-699-put-a-voice-agent-on-your-website-that-answers-from-your-doc/
- To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/