ERPNext AI Chat 🤖

An intelligent AI-powered chat assistant for ERPNext that uses LangChain and OpenAI GPT-4 to help users query and interact with their ERP data using natural language.

✨ Features

• 🗣️ Natural Language Queries - Ask questions in plain English
• 🎤 Voice Recognition - Speak your queries instead of typing! (NEW!)
• 🔧 7 Specialized Tools - Query customers, items, sales orders, purchase orders, stock, and more
• 🧠 Context-Aware - Remembers conversation history for multi-turn dialogues
• 🔐 Permission-Based - Respects ERPNext user permissions
• 💬 Real-Time Chat UI - Integrated directly into ERPNext navbar
• 📊 Session Management - Create, save, and manage chat sessions
• 🎯 Agentic AI - Automatically selects and executes the right tools

🚀 Quick Start

Prerequisites

• ERPNext 15.x or later
• Python 3.10+
• OpenAI API key (Get one here)

Installation

1. Get the app:

   cd /path/to/your/bench
   bench get-app https://github.com/barathprathosh/erpnext_ai_chat.git

2. Install on your site:

   bench --site your-site install-app erpnext_ai_chat

3. Install dependencies:

   bench pip install langchain langchain-openai langchain-community chromadb openai tiktoken

4. Configure API Key:

   • Go to: Setup → AI Chat Settings
   • Enter your OpenAI API key
   • Click Save

5. Restart bench:

   bench restart

6. Start chatting!

   • Click the "AI Assistant" button in the navbar
   • Type or click 🎤 to use voice input
   • Ask: "Show me pending sales orders"

💡 Example Queries

🎤 Voice or Type - Your Choice!

Click the microphone button and speak, or type your queries:

Sales & Orders

"Show me pending sales orders"
"Get sales orders for customer ABC Corp"
"Find completed orders from last week"

Customer Management

"Search for customer Tech Solutions"
"Get details for customer CUST-00001"
"Find all customers in California"

Inventory

"Check stock balance for ITEM-001"
"What's the available quantity in Main Warehouse?"
"Show me items with low stock"

Purchases

"Show pending purchase orders"
"Get POs from supplier XYZ"
"Find recent purchase orders"

🛠️ Architecture

User Query → Chat UI → API → AI Agent → LangChain → Tools → ERPNext Data
                                  ↓
                            OpenAI GPT-4
                                  ↓
                         Natural Language Response

Components

• AI Agent - Orchestrates conversation flow using LangChain
• 7 Specialized Tools - Query different ERPNext doctypes
• Memory Manager - Maintains conversation context
• REST API - 6 endpoints for chat operations
• Chat UI - Modern interface integrated into ERPNext

📚 Documentation

• Quick Start Guide - Get started in 5 minutes
• Setup Guide - Detailed installation and configuration
• Voice Feature Guide - Using voice recognition (NEW!)
• Project Structure - Technical architecture and code organization

🔧 Available Tools

Tool	Description	Example
search_customers	Find customers by name	"Search for customer ABC"
get_customer_details	Get detailed customer info	"Show details for CUST-001"
search_items	Find products/items	"Search for laptop items"
get_sales_orders	Fetch sales orders	"Show pending sales orders"
get_purchase_orders	Fetch purchase orders	"Show purchase orders from supplier X"
get_stock_balance	Check inventory levels	"Check stock for ITEM-001"
search_doctype	Generic doctype search	"Find all draft quotations"

🔌 API Endpoints

# Send a message
POST /api/method/erpnext_ai_chat.api.chat.send_message
{
    "message": "Show me pending sales orders",
    "session_id": "optional-session-id"
}

# Get chat history
GET /api/method/erpnext_ai_chat.api.chat.get_chat_history?session_id=AICS-0001

# Get all sessions
GET /api/method/erpnext_ai_chat.api.chat.get_sessions

# Create new session
POST /api/method/erpnext_ai_chat.api.chat.create_new_session

# Clear history
POST /api/method/erpnext_ai_chat.api.chat.clear_chat_history

# Delete session
POST /api/method/erpnext_ai_chat.api.chat.delete_session

💰 Cost

Uses OpenAI's API with affordable pricing:

Model	Cost per Query	Recommended For
gpt-4o-mini (default)	~$0.0001	Daily use (very cheap!)
gpt-4o	~$0.01	Complex queries

Example: 100 queries/day with gpt-4o-mini = ~$3/month

🔒 Security

• ✅ Respects ERPNext permissions
• ✅ API key stored encrypted
• ✅ All queries logged
• ✅ No direct database access
• ✅ User-based session isolation

🎯 Use Cases

• Customer Service - Quickly find customer information
• Sales Operations - Monitor orders and pipelines
• Inventory Management - Check stock levels instantly
• Procurement - Track purchase orders
• General Queries - Ask anything about your ERP data

🛠️ Extending

Add a Custom Tool

# In erpnext_ai_chat/ai_agent/tools.py

from langchain.tools import tool
import frappe

@tool
def get_invoices(status: str = None, limit: int = 10) -> str:
    """Get sales invoices with optional status filter"""
    invoices = frappe.get_all(
        "Sales Invoice",
        filters={"status": status} if status else {},
        fields=["name", "customer", "grand_total"],
        limit=limit
    )
    
    result = f"Found {len(invoices)} invoice(s):\n"
    for inv in invoices:
        result += f"- {inv.name}: {inv.customer} (${inv.grand_total})\n"
    
    return result

# Add to get_erpnext_tools() function
def get_erpnext_tools(user=None):
    return [
        # ... existing tools ...
        get_invoices  # Your new tool
    ]

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

MIT License - see LICENSE file for details

🙏 Acknowledgments

• Built with LangChain
• Powered by OpenAI GPT-4
• For Frappe ERPNext

📞 Support

• 📖 Documentation
• 🐛 Issue Tracker
• 💬 Discussions

⭐ Show Your Support

If you find this project useful, please consider giving it a star on GitHub!

---

Made with ❤️ for the ERPNext community