Receptionist AI AGENT

Overview

This project implements a Domain-Specific AI Intake Agent using the OpenMic API.
The agent acts as a Receptionist and demonstrates:

• Pre-call webhooks (fetch data before the call)
• In-call function calls (fetch details dynamically during the call)
• Post-call webhooks (log and process call summaries after the call)

---

Objective

• An intake agent that interacts naturally with users.
• Integrate OpenMic API with custom webhooks.
• Support Bot CRUD operations via UI.
• Demonstrate call flows (pre-call, in-call, post-call).

---

Features

• 🔹 Bot Management: Create, update, delete, and list bots.
• 🔹 Webhook Support:
  • Pre-call: Returns visitor info.
  • In-call: Fetches dynamic details using custom function calls.
  • Post-call: Stores summaries, transcripts, and follow-up actions.
• Logs & History: Displays call metadata, transcripts, and API call results.

---

Tech Stack

• Frontend: Next.js / React
• Backend: Node.js (Express routes)
• API Integration: OpenMic API
• Tools: ngrok (for exposing local endpoints)

---

🔗 Webhook & Function Flow

Pre-Call Webhook

• Returns visitor info before the call.

In-Call Function Call

• Agent requests an ID ( Employee Name).
• Fetches details in real time via API.

Post-Call Webhook

• Logs transcript, call summary, and metadata.
• Example: Follow-up request

---

Setup & Installation

# Clone repository
git clone https://github.com/Dipxssi/Receptionist-agent
cd Receptionist-agent

# Install dependencies
npm install

# Run development server
npm run dev

Expose local endpoints with ngrok:

ngrok http 3000

---

Receptionist Agent

• Pre-call: Returns visitor details.
• In-call: Finds employee location by name.
• Post-call: Logs visit details.

---