Product Agent

An intelligent, conversational planning system that turns ideas into professional specifications.

The Product Agent acts as an expert interviewer—covering Product, Design, Privacy, and Engineering—to guide you through a focused dialogue. As you chat, it autonomously synthesizes your inputs into a live set of structured documents:

• Product Requirements Document (PRD)
• User Experience Document (UXD) (Atomic Design)
• Privacy Design Document (PDD)
• Engineering Design Document (EDD)

Architecture

The system uses a multi-agent architecture orchestrated by Google Genkit:

1. The Analyst: Extracts facts and detects contradictions from user input.
2. The Synthesizer: Updates the four living documents with new information.
3. The Strategist: Analyzes the current state to ask the single most effective next question.

Tech Stack

• Framework: Next.js 15 (App Router)
• Language: TypeScript
• AI: Google Genkit + Gemini Models
• UI: React, Tailwind CSS, Radix UI

Getting Started

Prerequisites

• Node.js & npm
• A Google Gemini API Key

Installation

1. Clone the repo:

   git clone https://github.com/nan-bit/product-agent-app.git
   cd product-agent-app

2. Install dependencies:

   npm install

3. Configure Environment: Create a .env file in the root:

   GOOGLE_API_KEY="YOUR_GEMINI_API_KEY"

4. Run:

   npm run dev

   Open http://localhost:3000.

License

MIT