MCP Server

v 17.0 Third Party 212 468

Availability	Odoo Online Odoo.sh On Premise
Odoo Apps Dependencies	Discuss (mail)
Lines of code	1761
Technical Name	`mcp_server`
License	See License tab
Website	https://muchconsulting.de/
Versions	16.0 17.0 18.0 19.0

You bought this module and need support? Click here!

Availability	Odoo Online Odoo.sh On Premise
Odoo Apps Dependencies	Discuss (mail)
Lines of code	1761
Technical Name	`mcp_server`
License	See License tab
Website	https://muchconsulting.de/
Versions	16.0 17.0 18.0 19.0

Connect AI Assistants to Your Odoo Data with MCP Server

Enable natural language to search, create, update, and manage your Odoo records.

MCP Server for Odoo seamlessly connects any AI assistant that supports the Model Context Protocol to your Odoo instance, giving you the power to search, create, update, and manage records using natural language; all through a secure, standards-based integration.

What is Model Context Protocol (MCP)?

Model Context Protocol (MCP) is an open standard developed by Anthropic that enables seamless integration between AI assistants and data sources. It provides a universal way for AI systems to securely access and interact with local and remote resources while maintaining user control and privacy. One integration, every MCP-compatible AI tool.

🔌

Universal Compatibility

One integration works with ALL AI assistants that support MCP - no need for multiple custom integrations or proprietary connectors.

🔒

Secure by Design

Built-in security with controlled access, API key authentication, and granular per-model permissions for every operation.

🚀

Future-Proof

As new AI tools emerge, they'll support MCP - your Odoo integration automatically works with them. No vendor lock-in, ever.

Key Features & Benefits

🚀

Instant AI Integration

Connect any MCP-compatible AI assistant to your Odoo data without complex API development. One integration, every tool.

👥

Empower Your Team

Let employees use their preferred AI tools to access Odoo data naturally and efficiently - no retraining required.

</>

No Vendor Lock-in

Switch between AI providers anytime - your MCP integration continues to work seamlessly with any future MCP-compatible tool.

💬

Natural Language Operations

Not just queries - create, update, and delete records naturally. Ask "Create a new customer for Acme Corp" or "Update John Doe's phone number" and watch it happen instantly.

🛡️

Secure & Controlled Access

Control exactly which models AI can access with granular permissions. Secure API key authentication ensures your data remains protected at all times.

⚙️

Easy Configuration

Set up in minutes through Odoo's standard settings interface. No technical expertise required - just point, click, and configure.

Full CRUD Capabilities

Go beyond simple queries. Create, Read, Update, and Delete operations for any enabled model - all through natural language.

➕ Create

"Create a new customer for Acme Corp"

"Add a product called Premium Widget"

"Schedule a meeting for tomorrow"

🔍 Read

"Show me all customers from Spain"

"Find unpaid invoices from last month"

"List products below reorder point"

✏️ Update

"Update John's phone number"

"Change order status to confirmed"

"Set product price to $99.99"

🗑️ Delete

"Remove the test contact"

"Delete cancelled orders"

"Clean up duplicate records"

Works with ANY MCP-Compatible AI Assistant

The beauty of open standards - one integration, endless possibilities. MCP is rapidly becoming the industry standard, with new clients added regularly. View the growing list of compatible clients at modelcontextprotocol.io/clients.

Popular MCP Clients Include:

Claude Desktop

Anthropic's flagship MCP client

Cursor

AI-first code editor

Windsurf

Agentic IDE with MCP directory

VS Code

Native MCP in Copilot agent mode

And Many More...

🤖 Coding agents & CLIs: Claude Code, Codex CLI, Gemini CLI, Goose, Cline
💻 IDEs & editors: Zed, Continue.dev, Roo Code, Kilo Code
💬 Chat clients: ChatGPT (via Apps SDK), LibreChat, Cherry Studio
🧩 Frameworks & custom solutions: Chainlit, LangChain, any tool built with the MCP SDK

The MCP ecosystem now spans 500+ clients and is growing rapidly - new clients are added every week.

Feature Details

Dive deeper into the technical capabilities that make MCP Server for Odoo so powerful and flexible.

MCP Models Configuration in Odoo Settings

MCP Server Configuration in Odoo Settings

Full CRUD Through Natural Language

Complete data management through conversation, with the same ease as asking questions.

Create: Add new records to any enabled model - customers, products, orders, tasks, and more
Read: Search, filter, and retrieve records with complex criteria expressed in plain English
Update: Modify individual fields or bulk-update records based on conditions
Delete: Remove records with full respect for Odoo's referential integrity rules
Per-model toggles: Enable each operation independently for each exposed model

Enterprise-Grade Security

Security-first design with controls at every layer of the integration.

API key authentication: Token-based authentication with user-level permissions - no passwords stored
Easy key rotation: Generate, revoke, and rotate API keys directly from user settings
Audit logging: Comprehensive trail of every MCP operation for security monitoring and compliance
Model whitelist: Explicitly enable each model - default-deny posture protects unconfigured data
Encrypted storage: Credentials and tokens stored using Odoo's secure storage mechanisms

Granular Permission System

Control exactly what each AI assistant can do, down to the model and operation level.

Model-level permissions: Choose which Odoo models AI can access - res.partner, sale.order, account.move, or any custom model
Operation-level permissions: Independently enable Create, Read, Update, and Delete per model
Read-only mode: Grant safe reporting access without write capabilities
Full CRUD mode: Enable complete data management for trusted workflows
Familiar interface: All managed through standard Odoo settings - no separate admin tools

Multiple Transport Options

Flexible connectivity options to fit any deployment scenario.

stdio transport: Default option for local clients like Claude Desktop and Cursor
streamable-http transport: Web-based and remote clients with full bidirectional streaming
REST API endpoints: Standard HTTP integration for custom applications
XML-RPC compatibility: Works alongside Odoo's existing integration protocols
MCP client package: Drop-in mcp-server-odoo package for instant client setup

Built for Performance

Production-ready performance features for reliable operation at scale.

Connection pooling: Efficient reuse of database connections under high concurrency
Response caching: TTL-based caching reduces redundant queries for frequently-accessed data
Rate limiting: Built-in protection prevents runaway AI agents from overwhelming the server
Optimized payloads: Data formatting tuned for LLM context efficiency
Async-friendly: Designed to handle long-running AI sessions without blocking

Simple Setup Process

Get started in just 4 easy steps - from installation to your first AI-powered Odoo query in under 15 minutes.

Install Module

Install MCP Server in your Odoo instance like any other module.

Configure Access

Choose which models to expose and set permissions through Settings.

Generate API Key

Create secure API keys for authentication in user settings.

Connect AI

Install MCP client on your AI assistant and start querying.

Use Cases & Examples

🎧

Customer Service Acceleration

Scenario: Support agents need instant access to customer data while on calls
With MCP Server, support teams can ask their AI assistant to:

"Look up all orders for customer X in the last 6 months"
"Create a new support ticket for this complaint"
"Update the customer's preferred contact method to email"
"Show me unresolved tickets assigned to my team"
"Add a note about the resolution to this customer's record"

Business Impact: Agents handle calls without switching screens, reducing call times and improving customer satisfaction scores.

📈

Sales Intelligence on Demand

Scenario: Sales reps need to manage their CRM during prospect research and outreach
Sales teams can interact with their pipeline conversationally:

"Create a lead for Acme Corp with priority high"
"Update opportunity 'Q3 Enterprise Deal' status to won"
"Show me all opportunities closing this month"
"Which deals haven't had activity in 14+ days?"
"Add a follow-up activity for tomorrow on this lead"

Business Impact: Sales reps spend less time on data entry and more time selling, with cleaner pipeline data as a side effect.

📦

Inventory Management Made Conversational

Scenario: Warehouse managers need quick stock insights and updates
Operations teams can manage inventory through natural language:

"List products below reorder point"
"Create a new product variant for Widget Pro in red"
"Update stock levels for SKU-1234 to 500 units"
"Show me incoming shipments expected this week"
"Find products with no movement in the last 90 days"

Business Impact: Faster stock decisions, fewer stockouts, and reduced time spent navigating inventory screens.

💰

Financial Operations Through Chat

Scenario: Finance teams need rapid invoice processing and customer payment insights
Accountants can handle daily finance operations conversationally:

"Find unpaid invoices from last month over €5,000"
"Create an invoice for customer X with these line items"
"Update payment terms on this order to Net 60"
"Show overdue receivables grouped by customer"
"List all credit notes issued this quarter"

Business Impact: Month-end close runs faster, dunning workflows become proactive, and AR teams spot issues earlier.

👥

HR Records at Your Fingertips

Scenario: HR teams handle frequent record updates and lookups
HR professionals can manage employee data through conversation:

"Add new employee John Doe to the Engineering department"
"Update Sarah Chen's department to Marketing"
"List employees with upcoming work anniversaries"
"Show me all open positions in the company"
"Find employees who haven't completed onboarding"

Business Impact: Reduces administrative overhead for HR staff and surfaces actionable insights from people data.

📋

Project Management Without the Clicks

Scenario: Project managers juggle tasks, milestones, and resource allocation across multiple projects
PMs can drive their projects through natural language:

"Create a task 'Website redesign kickoff' due next Friday"
"Update milestone deadline for the Q4 launch to December 15"
"Show overdue tasks for the Marketing team"
"List all tasks blocked waiting on the design team"
"Assign this task to the next available developer"

Business Impact: PMs spend more time managing people and outcomes, less time wrangling tools.

Technical Specifications

Requirements

Odoo 17.0
MCP client package: mcp-server-odoo
Compatible AI assistant with MCP support

Protocols Supported

Model Context Protocol (MCP)
stdio transport (local clients)
streamable-http transport (web clients)
REST API endpoints
XML-RPC integration

Performance Features

Connection pooling for efficiency
Response caching with TTL
Rate limiting protection
Optimized data formatting for LLMs

Security Features

API key authentication
Per-model whitelist
Per-operation permissions
Complete audit logging

Ready to connect AI to your Odoo data?

Join the AI revolution and make your Odoo data instantly accessible through natural language. Experience the future of business data interaction today.

For support or questions, contact much. Consulting.

MCP Server for Odoo

Overview

The MCP Server module enables AI assistants to securely access and interact with your Odoo data through the Model Context Protocol (MCP). This module provides the server-side infrastructure within Odoo, while a separate Python package handles the MCP client communication. Supports full CRUD operations - create, read, update, and delete records through natural language.

Installation

Download and install the module in your Odoo instance
Navigate to Settings > MCP Server to configure access
Install UV on your local computer (where your AI assistant runs):

macOS/Linux:
```
curl -LsSf https://astral.sh/uv/install.sh | sh
```
Windows:
```
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
```
After installation, restart your terminal.
Configure your AI assistant to connect via MCP using the companion Python package

Configuration

Model Access

Go to Settings > MCP Server > Enabled Models
Add models you want to expose (e.g., res.partner, product.product)
Configure permissions for each model: - Read access - Write access - Create access - Delete access

API Key Setup

Navigate to Settings > Users & Companies > Users
Select the user for MCP access
Go to API Keys tab
Generate a new API key with appropriate scope
Use this key in your MCP client configuration

Client Setup

Important

The MCP Server runs on your local computer (where Claude Desktop or other AI assistants are installed), not on the Odoo server. The Python package connects from your local machine to your remote Odoo instance.

After installing the module and generating an API key, you can configure your AI assistant using either transport method:

Transport Options

The MCP server supports two transport types:

stdio transport (default): For local AI assistants like Claude Desktop, VS Code extensions
streamable-http transport: For web-based clients or remote connections

Standard Configuration (stdio transport):

{
  "mcpServers": {
    "odoo": {
      "command": "uvx",
      "args": ["mcp-server-odoo"],
      "env": {
        "ODOO_URL": "https://your-odoo-instance.com",
        "ODOO_API_KEY": "your-api-key-here"
      }
    }
  }
}

HTTP Transport Configuration:

For web-based or remote MCP clients:

{
  "mcpServers": {
    "odoo": {
      "command": "uvx",
      "args": ["mcp-server-odoo", "--transport", "streamable-http", "--port", "8000"],
      "env": {
        "ODOO_URL": "https://your-odoo-instance.com",
        "ODOO_API_KEY": "your-api-key-here"
      }
    }
  }
}

Then connect your client to http://localhost:8000/mcp/

Client-Specific Examples

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json using the configuration examples above.

Cursor

Add to ~/.cursor/mcp_settings.json using the configuration examples above.

Claude Code

Run the following command to add the Odoo MCP server:

claude mcp add odoo \
  -e ODOO_URL=https://your-odoo-instance.com \
  -e ODOO_API_KEY=your-api-key-here \
  -- uvx ivnvxd/mcp-server-odoo

Environment Variables

The MCP client requires one of the following authentication methods:

API Key Authentication (Recommended):

ODOO_URL: Your Odoo instance URL (e.g., https://mycompany.odoo.com)
ODOO_API_KEY: The API key generated in the previous step

Username/Password Authentication:

ODOO_URL: Your Odoo instance URL (e.g., https://mycompany.odoo.com)
ODOO_USER: Your Odoo username
ODOO_PASSWORD: Your Odoo password

Optional Variables:

ODOO_DB: Database name (auto-detected if not specified)

Security Groups

The module creates two security groups:

MCP Administrator: Can configure MCP settings and manage enabled models
MCP User: Can access MCP-enabled models based on configured permissions

Usage Examples

Once configured, you can query and manage your Odoo data using natural language:

Data Retrieval:

"Show me all customers from Spain"
"Find products with stock below 10 units"
"List today's sales orders over $1000"
"Search for unpaid invoices from last month"

Data Management:

"Create a new customer contact for Acme Corporation"
"Add a new product called 'Premium Widget' with price $99.99"
"Update the phone number for customer John Doe"
"Change the status of order SO/2024/001 to confirmed"
"Delete the test contact we created earlier"

API Endpoints

The module provides several REST and XML-RPC endpoints:

REST API

/mcp/health - Health check
/mcp/system/info - System information
/mcp/auth/validate - API key validation
/mcp/models - List enabled models
/mcp/models/{model}/access - Check model permissions

XML-RPC API

/mcp/xmlrpc/common - Authentication
/mcp/xmlrpc/db - Database operations
/mcp/xmlrpc/object - Model operations

Security Considerations

Use HTTPS in production environments
Generate unique API keys for each integration
Configure model access carefully - only enable necessary models
Regularly review audit logs for suspicious activity
Keep the module updated

Troubleshooting

Common Issues

Module Not Installing - Check that all dependencies are satisfied - Ensure Odoo 17.0 is being used

API Key Not Working - Verify the key is active in user settings - Check user has appropriate MCP permissions - Ensure correct API key scope

Model Access Denied - Confirm model is in enabled models list - Check operation permissions for the model - Verify user's security group membership

"spawn uvx ENOENT" Error

This error means UV is not installed on your local computer:

Install UV using the commands in the Installation section above
Restart your terminal and Claude Desktop
On macOS, if the issue persists, launch Claude from Terminal:
```
open -a "Claude"
```
Alternative: Use the full path to uvx (find it with which uvx)

Database Access Denied

If you see "Access Denied" when the MCP server tries to list databases:

This is normal security behavior on some Odoo instances
You must specify the ODOO_DB environment variable in your configuration
The server will use your specified database without validation

Support

For support and documentation:

GitHub Repository: https://github.com/ivnvxd/mcp-server-odoo

much. Consulting License v1.0

This software and associated files (the “Software”) can only be used (executed, modified, executed after modifications) with a valid contract or licensing (with much. Consulting GmbH) and for the correct number of users.

With a valid Partnership Agreement with much. Consulting GmbH, the above permissions are also granted, as long as the usage is limited to a testing or development environment.

You may develop Odoo modules based on the Software and distribute them under the license of your choice, provided that it is compatible with the terms of the much. Consulting License (For example: Commercial licenses, or proprietary licenses similar to this one).

You may use Odoo modules published under any license along with the Software, provided that their license is compatible with the terms of the much. Consulting GmbH License and the Odoo Enterprise License (Including, but not limited to, any module published on the Odoo Apps Store)

It is forbidden to publish, distribute, sublicense, or sell copies of the Software or modified copies of the Software.

The above copyright notice and this permission notice must be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON INFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Please log in to comment on this module

The author can leave a single reply to each comment.
This section is meant to ask simple questions or leave a rating. Every report of a problem experienced while using the module should be addressed to the author directly (refer to the following point).
If you want to start a discussion with the author, please use the developer contact information. They can usually be found in the description.

Ratings
Discuss

There are no ratings yet!

Demo Server Request

webshop.admin@zafco.com

on 4/24/26, 4:49 AM Confirmed Purchase

Hello Andrey,

Can you share demo link for us to check it.

Download Invoice

NetConnections GmbH

on 4/13/26, 12:49 PM

Is it possible to search for outgoing invoices via the MCP server and download them as PDF files?

Download Invoice

Andrey Ivanov

on 4/14/26, 6:31 AM

Yes and no.

Searching invoices works fine — just enable the Invoices model (account.move) with read access in the MCP Server settings, and your AI assistant can query them.

Downloading PDFs isn't supported yet. The MCP Server only exposes standard data operations (read/search/create/update/delete), not Odoo's report/print engine, so invoice PDFs can't be generated on demand.

One partial workaround: if an invoice has already been printed in Odoo, its PDF is stored as an attachment and can be fetched via the Attachments model. It won't generate a new one, just return what's already there.

On-demand PDF export is a good feature request — I've noted it.

Jatin Vamja

on 11/24/25, 10:03 PM Confirmed Purchase

I have installed module MCP module in ODOO 17 enterprise

I am following these steps. I have done step 1 and 2 but can not find API Keys tab inside user form view.
How would I generate API key and where should enter in ODOO ?

Navigate to Settings > Users & Companies > Users
Select the user for MCP access
Go to API Keys tab
Generate a new API key with appropriate scope
Use this key in your MCP client configuration