HR Policy Generation Agent

AI-powered platform for creating, managing, and maintaining compliant HR policies

Overview

The HR Policy Generation Agent is an intelligent platform that leverages AI to streamline the creation, updating, and management of HR policies for organizations. Built with Next.js and powered by Lyzr AI agents, it provides an intuitive interface for generating compliant, country-specific HR policies through conversational AI.

Key Features

AI-Powered Policy Generation - Create comprehensive HR policies through natural conversation with specialized AI agents
Multi-Country Compliance - Generate policies compliant with regulations across multiple jurisdictions (US, Canada, UK, Germany, France, India, Australia)
Intelligent Policy Updates - Update existing policies with AI-assisted compliance checking and version control
PDF Upload & Analysis - Upload existing policy documents for analysis, parsing, and conversion to editable format
Policy Database - Centralized repository for storing, viewing, and managing all organizational policies
Real-time Research - Integrated Perplexity AI agent for up-to-date legal and compliance research
Rich Text Editing - Professional editor with formatting, highlighting, and markdown support
Version Control - Track policy changes with versioning and status management (Draft, In Review, Approved, Published, Archived)
Dark Mode Support - Comfortable viewing experience with theme toggling

Architecture

The application is built using a modern stack with AI agents at its core:

┌─────────────────────────────────────────────────────────────┐
│                      Next.js Frontend                        │
│  (React 19, TypeScript, Tailwind CSS, shadcn/ui)           │
└───────────────────┬─────────────────────────────────────────┘
                    │
┌───────────────────┴─────────────────────────────────────────┐
│                   API Routes Layer                           │
│  (/api/chat, /api/policies, /api/upload-pdf, etc.)         │
└───────────────────┬─────────────────────────────────────────┘
                    │
        ┌───────────┴───────────┐
        │                       │
┌───────┴────────┐     ┌────────┴─────────┐
│   Lyzr AI      │     │    MongoDB        │
│   Agents       │     │    (Prisma)       │
│                │     │                   │
│ • Chat Agent   │     │ • Users           │
│ • Perplexity   │     │ • Policies        │
│ • Compliance   │     │ • Sessions        │
│ • Law Updates  │     └───────────────────┘
└────────────────┘

Getting Started

Prerequisites

Node.js 18+
npm, yarn, pnpm, or bun
MongoDB database
Lyzr AI API key (Get one here)

Installation

Clone the repository

git clone https://github.com/amitlyzr/policy-generation-app.git
cd policy-generation-app

Install dependencies

npm install
# or
yarn install
# or
pnpm install

Set up environment variables

Create a .env.local file in the root directory:

# Database
DATABASE_URL="your_mongodb_connection_string"

# Lyzr AI Configuration
LYZR_API_KEY="your_lyzr_api_key"

# Agent IDs (obtain from Lyzr dashboard)
NEXT_PUBLIC_CHAT_AGENT_ID="your_chat_agent_id"
NEXT_PUBLIC_PERPLEXITY_AGENT_ID="your_perplexity_agent_id"
NEXT_PUBLIC_COMPLIANCE_AGENT_ID="your_compliance_agent_id"
NEXT_PUBLIC_LAW_AGENT_ID="your_law_agent_id"

Note

You can find a template in .env.example. Copy it and fill in your values.

Set up the database

npx prisma generate
npx prisma db push

Run the development server

npm run dev
# or
yarn dev
# or
pnpm dev

Open http://localhost:3000 to see the application.

Usage

Creating a New Policy

Navigate to Create New Policy from the sidebar
Start a conversation with the AI agent about the policy you need
Specify the policy type (e.g., Remote Work, PTO, Code of Conduct)
Select applicable countries for compliance
Review and refine the generated policy through conversation
Save the policy to your database

Uploading Existing Policies

Go to Upload Policy section
Drag and drop or select a PDF file
Add a policy name and select applicable countries
The system will parse and convert the PDF to editable markdown format
Review and save to your policy database

Updating Policies

Navigate to Compliance Test section
Select a policy from your database
Describe the changes needed or compliance requirements
The AI will analyze and suggest updates
Review changes and save new version

Managing Policy Database

Access Policy Database to view all policies
Filter by status, country, or search by name
View, edit, or archive policies as needed
Track version history and status changes

Technology Stack

Frontend

Next.js 15.5 - React framework with App Router
React 19 - UI library
TypeScript - Type safety
Tailwind CSS 4 - Utility-first CSS
shadcn/ui - Component library
Radix UI - Unstyled accessible components

AI & Processing

Lyzr AI SDK - Multi-agent orchestration
Tiptap - Rich text editor
pdf2md - PDF parsing
Marked - Markdown processing
Streamdown - Streaming markdown renderer

Backend & Database

Prisma - Database ORM
MongoDB - Document database
Next.js API Routes - Backend API

State Management & Auth

Zustand - Lightweight state management
Lyzr Auth - Authentication provider
js-cookie - Cookie management

Project Structure

policy-generation-agent/
├── prisma/
│   └── schema.prisma           # Database schema
├── public/                     # Static assets
├── src/
│   ├── app/
│   │   ├── api/               # API routes
│   │   │   ├── chat/          # Chat agent endpoint
│   │   │   ├── policies/      # Policy CRUD operations
│   │   │   ├── upload-pdf/    # PDF processing
│   │   │   └── research-policy/ # Research agent endpoint
│   │   ├── create-policy/     # Policy creation page
│   │   ├── update-policy/     # Policy update page
│   │   ├── upload-policy/     # Policy upload page
│   │   ├── policy-db/         # Policy database page
│   │   ├── layout.tsx         # Root layout
│   │   └── page.tsx           # Dashboard
│   ├── components/
│   │   ├── custom/            # Custom components
│   │   │   ├── policy-conversation.tsx
│   │   │   ├── policy-database.tsx
│   │   │   ├── policy-uploader.tsx
│   │   │   └── ...
│   │   ├── providers/         # Context providers
│   │   └── ui/                # Reusable UI components
│   ├── lib/
│   │   ├── lyzr.ts           # Lyzr AI integration
│   │   ├── prisma.ts         # Prisma client
│   │   ├── file-utils.ts     # File processing utilities
│   │   └── utils.ts          # General utilities
│   └── stores/
│       └── policy-store.ts   # Zustand state management
├── .env.example              # Environment variables template
├── package.json
└── README.md

API Documentation

Chat Endpoint

POST /api/chat
{
  "message": "Create a remote work policy",
  "sessionId": "session_xyz",
  "agentId": "chat_agent_id",
  "userId": "user_123"
}

Policy Management

// Get all policies for a user
GET /api/policies?userId=user_123

// Create new policy
POST /api/policies
{
  "name": "Remote Work Policy",
  "content": "...",
  "countries": ["United States", "Canada"],
  "userId": "user_123"
}

// Update policy
PUT /api/policies?policyId=policy_123

// Delete policy
DELETE /api/policies?policyId=policy_123

PDF Upload

POST /api/upload-pdf
Content-Type: multipart/form-data
{
  "file": <PDF file>,
  "userId": "user_123"
}

Configuration

Lyzr AI Agents

The application uses multiple specialized AI agents:

Chat Agent - Main conversational interface for policy generation
Perplexity Agent - Real-time research for legal and compliance information
Compliance Agent - Validates policies against regulatory requirements
Law Agent - Provides updates on legal changes and requirements

Configure these agents in the Lyzr dashboard and add their IDs to your environment variables.

Database Schema

model User {
  id        String   @id @default(auto()) @map("_id") @db.ObjectId
  email     String   @unique
  userId    String   @unique
  orgId     String
  policies  Policy[]
}

model Policy {
  id          String       @id @default(auto()) @map("_id") @db.ObjectId
  policyId    String       @unique
  name        String
  content     String
  countries   String[]
  status      PolicyStatus @default(DRAFT)
  version     Int          @default(1)
  userId      String       @db.ObjectId
  user        User         @relation(fields: [userId], references: [id])
}

enum PolicyStatus {
  DRAFT
  IN_REVIEW
  APPROVED
  PUBLISHED
  ARCHIVED
}

Development

Running Tests

npm test

Building for Production

npm run build
npm start

Linting

npm run lint

Deployment

Vercel (Recommended)

The easiest way to deploy is using Vercel:

Push your code to GitHub
Import project to Vercel
Add environment variables
Deploy

Other Platforms

The application can be deployed to any platform that supports Next.js:

AWS Amplify
Netlify
Railway
DigitalOcean App Platform

Ensure environment variables are properly configured in your deployment platform.

Troubleshooting

Common Issues

Database Connection Errors

Verify your DATABASE_URL is correctly formatted
Ensure MongoDB is accessible from your deployment environment
Run npx prisma generate after schema changes

AI Agent Errors

Verify your LYZR_API_KEY is valid
Check that all agent IDs are correctly configured
Ensure you have sufficient credits in your Lyzr account

PDF Upload Issues

Check file size (max 10MB by default)
Verify PDF is not password-protected
Ensure proper file permissions

Support

Acknowledgments

Built with Lyzr AI
UI components from shadcn/ui
Icons from Tabler Icons

Made with ❤️ by the Lyzr team
_{Empowering organizations with AI-driven HR policy management}

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
prisma		prisma
public		public
src		src
.env.example		.env.example
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
components.json		components.json
eslint.config.mjs		eslint.config.mjs
next.config.ts		next.config.ts
package-lock.json		package-lock.json
package.json		package.json
postcss.config.mjs		postcss.config.mjs
tsconfig.json		tsconfig.json

License

amitlyzr/policy-generation-app-os

Folders and files

Latest commit

History

Repository files navigation