How to integrate Erpnext MCP with OpenAI Agents SDK

Framework Integration Gradient
Erpnext Logo
open-ai-agents-sdk Logo
divider

Introduction

This guide walks you through connecting Erpnext to the OpenAI Agents SDK using the Composio tool router. By the end, you'll have a working Erpnext agent that can list overdue tasks for all projects, create a new customer record, get all open purchase orders through natural language commands.

This guide will help you understand how to give your OpenAI Agents SDK agent real control over a Erpnext account through Composio's Erpnext MCP server.

Before we dive in, let's take a quick look at the key ideas and tools involved.

TL;DR

Here's what you'll learn:
  • Get and set up your OpenAI and Composio API keys
  • Install the necessary dependencies
  • Initialize Composio and create a Tool Router session for Erpnext
  • Configure an AI agent that can use Erpnext as a tool
  • Run a live chat session where you can ask the agent to perform Erpnext operations

What is open-ai-agents-sdk?

The OpenAI Agents SDK is a lightweight framework for building AI agents that can use tools and maintain conversation state. It provides a simple interface for creating agents with hosted MCP tool support.

Key features include:

  • Hosted MCP Tools: Connect to external services through hosted MCP endpoints
  • SQLite Sessions: Persist conversation history across interactions
  • Simple API: Clean interface with Agent, Runner, and tool configuration
  • Streaming Support: Real-time response streaming for interactive applications

What is the Erpnext MCP server, and what's possible with it?

The Erpnext MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Erpnext account. It provides structured and secure access so your agent can perform Erpnext operations on your behalf.

Supported Tools & Triggers

Tools
Add CommentTool to add a comment to a document in ERPNext/Frappe.
Add TagTool to add a tag to a document in ERPNext.
Apply WorkflowTool to apply a workflow action to a document in ERPNext/Frappe.
Cancel DocumentCancel a submitted document in ERPNext/Frappe to change its status from Submitted to Cancelled.
Create DocumentTool to create a new document of a specific DocType in ERPNext.
Create TimesheetTool to create a new Timesheet record in ERPNext.
Create WebhookTool to create a new webhook configuration in ERPNext.
Delete DocumentTool to delete a document using the Frappe client API.
Delete DocumentTool to delete a specific document by DocType and name.
Download file from ERPNextTool to download a file from ERPNext by its URL.
Download PDF DocumentTool to download a document as PDF from ERPNext with optional print format.
Get All LanguagesTool to get a list of all available languages in the ERPNext/Frappe system.
Get All RolesTool to get a list of all roles available in the ERPNext system.
Get DocumentTool to get a single document by DocType and name or filters from Frappe/ERPNext.
Get Document CountTool to get the count of documents matching specified filters in ERPNext/Frappe.
Get DocType MetadataTool to retrieve complete DocType metadata/schema including field definitions, field types, permissions, and configurations.
Get DocumentTool to retrieve a specific document by its DocType and name (ID).
Get Document with MetadataTool to retrieve a document with full metadata including attachments, comments, activity logs, and related information.
Get Exchange RateTool to get the currency exchange rate between two currencies in ERPNext.
Get Fiscal YearTool to get fiscal year information for a given date in ERPNext.
Get Framework VersionTool to get the Frappe framework version and all installed app versions.
Get Item DetailsTool to get detailed item information including pricing, taxes, and stock details from ERPNext.
Get List of DocumentsTool to retrieve a list of documents from ERPNext/Frappe with filtering, field selection, and pagination.
Get Logged UserTool to get the email/ID of the currently authenticated user.
Get Party DetailsTool to get comprehensive customer or supplier details including addresses, contacts, and default financial settings.
Get Payment EntryTool to get payment entry details for an invoice or order from ERPNext.
Get Stock BalanceTool to retrieve the current stock balance for a specific item in a warehouse.
Get TimezonesTool to get a list of all available timezones in the ERPNext system.
Get User RolesTool to get roles assigned to a user.
Get Field ValueTool to get specific field value(s) from a document in ERPNext.
Get Workflow TransitionsTool to get available workflow transitions for a document.
Insert DocumentTool to insert a new document in ERPNext/Frappe using the client API.
Insert Multiple DocumentsTool to insert multiple documents at once into ERPNext/Frappe.
List DocTypesTool to get a list of all DocTypes available in the ERPNext system.
List ERPNext DocumentsTool to list documents of a specific DocType from ERPNext.
List EmployeesTool to retrieve a list of Employee records from ERPNext.
List ProjectsTool to retrieve a list of Project records from ERPNext.
List TimesheetsTool to get a list of Timesheet records from ERPNext.
List WebhooksTool to list webhook configurations in ERPNext.
Make Delivery NoteCreate a draft Delivery Note from an existing Sales Order in ERPNext.
Make Purchase OrderCreate a draft Purchase Order from an existing Material Request in ERPNext.
Make Sales InvoiceTool to create a Sales Invoice from an existing Sales Order in ERPNext.
Make Stock EntryTool to create a Stock Entry for material transfer, receipt, or issue in ERPNext.
Ping APITool to check if the ERPNext/Frappe API is reachable.
Rename DocumentTool to rename an ERPNext document by changing its unique ID/name.
Save Document with ActionTool to save, submit, cancel, or update a document in ERPNext.
Save DocumentTool to save an existing ERPNext/Frappe document with changes.
Global SearchTool to perform global text search across ERPNext documents.
Search Link Field DocumentsTool to search for documents to link in ERPNext/Frappe Link fields.
Set ValueTool to set a specific field value on a document in ERPNext.
Submit DocumentSubmit a draft document in ERPNext/Frappe to change its status from Draft to Submitted.
Update ERPNext documentTool to update a specific ERPNext document.

What is the Composio tool router, and how does it fit here?

What is Tool Router?

Composio's Tool Router helps agents find the right tools for a task at runtime. You can plug in multiple toolkits (like Gmail, HubSpot, and GitHub), and the agent will identify the relevant app and action to complete multi-step workflows. This can reduce token usage and improve the reliability of tool calls. Read more here: Getting started with Tool Router

The tool router generates a secure MCP URL that your agents can access to perform actions.

How the Tool Router works

The Tool Router follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Prerequisites

Before starting, make sure you have:
  • Composio API Key and OpenAI API Key
  • Primary know-how of OpenAI Agents SDK
  • A live Erpnext project
  • Some knowledge of Python or Typescript

Getting API Keys for OpenAI and Composio

OpenAI API Key
  • Go to the OpenAI dashboard and create an API key. You'll need credits to use the models, or you can connect to another model provider.
  • Keep the API key safe.
Composio API Key

Install dependencies

pip install composio_openai_agents openai-agents python-dotenv

Install the Composio SDK and the OpenAI Agents SDK.

Set up environment variables

bash
OPENAI_API_KEY=sk-...your-api-key
COMPOSIO_API_KEY=your-api-key
USER_ID=composio_user@gmail.com

Create a .env file and add your OpenAI and Composio API keys.

Import dependencies

import asyncio
import os
from dotenv import load_dotenv

from composio import Composio
from composio_openai_agents import OpenAIAgentsProvider
from agents import Agent, Runner, HostedMCPTool, SQLiteSession
What's happening:
  • You're importing all necessary libraries.
  • The Composio and OpenAIAgentsProvider classes are imported to connect your OpenAI agent to Composio tools like Erpnext.

Set up the Composio instance

load_dotenv()

api_key = os.getenv("COMPOSIO_API_KEY")
user_id = os.getenv("USER_ID")

if not api_key:
    raise RuntimeError("COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key")

# Initialize Composio
composio = Composio(api_key=api_key, provider=OpenAIAgentsProvider())
What's happening:
  • load_dotenv() loads your .env file so OPENAI_API_KEY and COMPOSIO_API_KEY are available as environment variables.
  • Creating a Composio instance using the API Key and OpenAIAgentsProvider class.

Create a Tool Router session

# Create a Erpnext Tool Router session
session = composio.create(
    user_id=user_id,
    toolkits=["erpnext"]
)

mcp_url = session.mcp.url

What is happening:

  • You give the Tool Router the user id and the toolkits you want available. Here, it is only erpnext.
  • The router checks the user's Erpnext connection and prepares the MCP endpoint.
  • The returned session.mcp.url is the MCP URL that your agent will use to access Erpnext.
  • This approach keeps things lightweight and lets the agent request Erpnext tools only when needed during the conversation.

Configure the agent

# Configure agent with MCP tool
agent = Agent(
    name="Assistant",
    model="gpt-5",
    instructions=(
        "You are a helpful assistant that can access Erpnext. "
        "Help users perform Erpnext operations through natural language."
    ),
    tools=[
        HostedMCPTool(
            tool_config={
                "type": "mcp",
                "server_label": "tool_router",
                "server_url": mcp_url,
                "headers": {"x-api-key": api_key},
                "require_approval": "never",
            }
        )
    ],
)
What's happening:
  • We're creating an Agent instance with a name, model (gpt-5), and clear instructions about its purpose.
  • The agent's instructions tell it that it can access Erpnext and help with queries, inserts, updates, authentication, and fetching database information.
  • The tools array includes a HostedMCPTool that connects to the MCP server URL we created earlier.
  • The headers dict includes the Composio API key for secure authentication with the MCP server.
  • require_approval: 'never' means the agent can execute Erpnext operations without asking for permission each time, making interactions smoother.

Start chat loop and handle conversation

print("\nComposio Tool Router session created.")

chat_session = SQLiteSession("conversation_openai_toolrouter")

print("\nChat started. Type your requests below.")
print("Commands: 'exit', 'quit', or 'q' to end\n")

async def main():
    try:
        result = await Runner.run(
            agent,
            "What can you help me with?",
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")
    except Exception as e:
        print(f"Error: {e}\n")

    while True:
        user_input = input("You: ").strip()
        if user_input.lower() in {"exit", "quit", "q"}:
            print("Goodbye!")
            break

        result = await Runner.run(
            agent,
            user_input,
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")

asyncio.run(main())
What's happening:
  • The program prints a session URL that you visit to authorize Erpnext.
  • After authorization, the chat begins.
  • Each message you type is processed by the agent using Runner.run().
  • The responses are printed to the console, and conversations are saved locally using SQLite.
  • Typing exit, quit, or q cleanly ends the chat.

Complete Code

Here's the complete code to get you started with Erpnext and open-ai-agents-sdk:

import asyncio
import os
from dotenv import load_dotenv

from composio import Composio
from composio_openai_agents import OpenAIAgentsProvider
from agents import Agent, Runner, HostedMCPTool, SQLiteSession

load_dotenv()

api_key = os.getenv("COMPOSIO_API_KEY")
user_id = os.getenv("USER_ID")

if not api_key:
    raise RuntimeError("COMPOSIO_API_KEY is not set. Create a .env file with COMPOSIO_API_KEY=your_key")

# Initialize Composio
composio = Composio(api_key=api_key, provider=OpenAIAgentsProvider())

# Create Tool Router session
session = composio.create(
    user_id=user_id,
    toolkits=["erpnext"]
)
mcp_url = session.mcp.url

# Configure agent with MCP tool
agent = Agent(
    name="Assistant",
    model="gpt-5",
    instructions=(
        "You are a helpful assistant that can access Erpnext. "
        "Help users perform Erpnext operations through natural language."
    ),
    tools=[
        HostedMCPTool(
            tool_config={
                "type": "mcp",
                "server_label": "tool_router",
                "server_url": mcp_url,
                "headers": {"x-api-key": api_key},
                "require_approval": "never",
            }
        )
    ],
)

print("\nComposio Tool Router session created.")

chat_session = SQLiteSession("conversation_openai_toolrouter")

print("\nChat started. Type your requests below.")
print("Commands: 'exit', 'quit', or 'q' to end\n")

async def main():
    try:
        result = await Runner.run(
            agent,
            "What can you help me with?",
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")
    except Exception as e:
        print(f"Error: {e}\n")

    while True:
        user_input = input("You: ").strip()
        if user_input.lower() in {"exit", "quit", "q"}:
            print("Goodbye!")
            break

        result = await Runner.run(
            agent,
            user_input,
            session=chat_session
        )
        print(f"Assistant: {result.final_output}\n")

asyncio.run(main())

Conclusion

This was a starter code for integrating Erpnext MCP with OpenAI Agents SDK to build a functional AI agent that can interact with Erpnext.

Key features:

  • Hosted MCP tool integration through Composio's Tool Router
  • SQLite session persistence for conversation history
  • Simple async chat loop for interactive testing
You can extend this by adding more toolkits, implementing custom business logic, or building a web interface around the agent.

How to build Erpnext MCP Agent with another framework

FAQ

What are the differences in Tool Router MCP and Erpnext MCP?

With a standalone Erpnext MCP server, the agents and LLMs can only access a fixed set of Erpnext tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Erpnext and many other apps based on the task at hand, all through a single MCP endpoint.

Can I use Tool Router MCP with OpenAI Agents SDK?

Yes, you can. OpenAI Agents SDK fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Erpnext tools.

Can I manage the permissions and scopes for Erpnext while using Tool Router?

Yes, absolutely. You can configure which Erpnext scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Erpnext data and credentials are handled as safely as possible.

Used by agents from

Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai
Context
Letta
glean
HubSpot
Agent.ai
Altera
DataStax
Entelligence
Rolai

Never worry about agent reliability

We handle tool reliability, observability, and security so you never have to second-guess an agent action.