mcp-odoo-adv
Enhanced Odoo MCP server (STDIO, SSE, Streamable HTTP)
Install / Use
claude mcp add AlanOgic -- npx -y github:AlanOgic/mcp-odoo-advIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of mcp-odoo-adv
mcp-odoo-adv scores 78/100 on our quality scale, 3713th of 4,702 Development & Engineering skills we index.
Its MCP Server is 16 KB long, well organised into 59 sections with 22 code examples: a thorough specification that gives an agent plenty to work with.
It has 10 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so mcp-odoo-adv is actively maintained.
- Our last check on 2026-11-13 found the source still online.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 80/100, with 2 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-09-25. Automated pattern scan on 2026-09-24. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
mcp-odoo-adv compared with similar skills
All 4 of these similar skills score higher than mcp-odoo-adv; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| mcp-odoo-adv (this skill)by AlanOgic | 78 | 10 | 2mo ago | MCP Server |
| taste-skillby Leonxlnx | 100 | 117.6k | 9d ago | CLAUDE.md |
| Agent-Reachby Panniantong | 100 | 112.7k | 9d ago | CLAUDE.md |
| rufloby ruvnet | 100 | 82.0k | today | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.9k | today | CLAUDE.md |
Frequently asked questions
- How do I install mcp-odoo-adv?
- Run
claude mcp add AlanOgic -- npx -y github:AlanOgic/mcp-odoo-adv. The install tabs above show the steps for each supported agent. - Which AI agents does mcp-odoo-adv work with?
- It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
- Is mcp-odoo-adv safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 80/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is mcp-odoo-adv still maintained?
- The repository was last updated about 2 months ago, so mcp-odoo-adv is actively maintained.
Skill content
View source on GitHubOdoo MCP Server Advanced
Two universal tools. A growing cookbook. Full Odoo API access for AI assistants.
An advanced MCP (Model Context Protocol) server for Odoo ERP — STDIO, SSE, and Streamable HTTP transports — that gives Claude, Cursor, ChatGPT-via-MCP and other AI hosts direct access to your Odoo data and workflows.
AI assistants: read AGENTS.md for the tool-first quick reference.
🎯 Philosophy: Simplicity & Power
Connect AI assistants to Odoo with two universal tools plus a self-improving knowledge base:
execute_method— call any Odoo method on any modelbatch_execute— execute multiple operations atomically (with@Nresult references)add_cookbook_pattern— document hard-won recipes after ≥4 failed attempts
Plus 11 discovery resources, 3 prompts, and an odoo://cookbook/patterns resource that surfaces past lessons. No specialized tools, no validation layer, no artificial limits — full Odoo API.
To rule them all. What You Can Do
Everything you need, just ask AI: automate, query, manage, customize, develop new modules, integrate with external systems, enhance your Odoo instance through AI.
🚀 Quick Start
Installation
Option 1: Traditional pip install
# From source
git clone https://github.com/AlanOgic/mcp-odoo-adv.git
cd mcp-odoo-adv
# Virtual Environment
python3 -m venv .venv
source .venv/bin/activate
# Installation
pip install -e .
Option 2: Using uvx (No Installation)
# From source directory
uvx --from . odoo-mcp
Option 3: Using Docker
# build STDIO
docker build -t alanogic/mcp-odoo-adv:latest -f Dockerfile .
# build SSE
docker build -t alanogic/mcp-odoo-adv-sse:latest -f Dockerfile.sse .
# build HTTP
docker build -t alanogic/mcp-odoo-adv-http:latest -f Dockerfile.http .
# Run STDIO
docker run --env-file .env alanogic/mcp-odoo-adv:latest
# Run SSE
docker run --env-file .env alanogic/mcp-odoo-adv-sse:latest
# Run HTTP
docker run --env-file .env alanogic/mcp-odoo-adv-http:latest
Configuration
Create a .env file (minimum):
cp .env.example .env
vim .env
ODOO_URL=https://your-odoo-instance.com
ODOO_DB=your-database-name
ODOO_USERNAME=your-username
ODOO_PASSWORD=your-password-or-api-key
Optional: Custom Configuration Directory
Organize multiple Odoo configurations by setting a custom directory:
# Create custom config directory
export ODOO_CONFIG_DIR=~/mcp-odoo-env
mkdir -p $ODOO_CONFIG_DIR
# Copy and configure
cp .env.example $ODOO_CONFIG_DIR/.env
vim $ODOO_CONFIG_DIR/.env
# Run server (automatically uses custom directory)
odoo-mcp
This is useful for:
- Managing multiple Odoo instances (dev, staging, production)
- Organizing configs outside project directory
- Docker/Compose deployments with volume mounts
Run Server
After pip install -e ., four console scripts are wired up:
# STDIO (Claude Desktop, Claude Code, Cursor)
odoo-mcp # or: python -m odoo_mcp
# SSE (browsers, port 8009)
odoo-mcp-sse
# Streamable HTTP (API integrations, port 8008)
odoo-mcp-http
# HTTP with Bearer auth (production behind a reverse proxy)
odoo-mcp-http-secure
Each runner streams logs to ./logs/mcp_server_<transport>_<timestamp>.log plus stderr.
Claude Desktop Setup
Option 1: Using local installation
Add to claude_desktop_config.json:
{
"mcpServers": {
"odoo": {
"command": "python",
"args": ["-m", "odoo_mcp"],
"env": {
"ODOO_URL": "https://your-instance.odoo.com",
"ODOO_DB": "your-database",
"ODOO_USERNAME": "your-username",
"ODOO_PASSWORD": "your-password"
}
}
}
}
Option 2: Using uvx (Recommended - No Installation)
First, create your credentials in ~/.config/odoo/.env:
mkdir -p ~/.config/odoo
cat > ~/.config/odoo/.env << 'EOF'
ODOO_URL=https://your-instance.odoo.com
ODOO_DB=your-database
ODOO_USERNAME=your-username
ODOO_PASSWORD=your-password
EOF
Then add to claude_desktop_config.json:
{
"mcpServers": {
"odoo": {
"command": "uvx",
"args": ["--from", "odoo-mcp", "odoo-mcp"]
}
}
}
Benefits:
- No installation required - uvx downloads and runs automatically
- Credentials stored securely in
.envfile (not in config) - Always uses the latest published version
- Works from anywhere on your system
Option 3: Using Docker
{
"mcpServers": {
"odoo": {
"command": "docker",
"args": [
"run", "-i", "--rm",
"-e", "ODOO_URL=https://your-instance.odoo.com",
"-e", "ODOO_DB=your-database",
"-e", "ODOO_USERNAME=your-username",
"-e", "ODOO_PASSWORD=your-password",
"alanogic/mcp-odoo-adv:latest"
]
}
}
}
Or with .env file:
{
"mcpServers": {
"odoo": {
"command": "docker",
"args": ["run", "-i", "--rm", "--env-file", "/absolute/path/to/.env", "alanogic/mcp-odoo-adv:latest"]
}
}
}
🔧 The Tools
1. execute_method - The Universal Powerhouse
Execute ANY Odoo method on ANY model. Full API access. No limitations.
execute_method(
model="<model_name>",
method="<method_name>",
args_json='[...]', # Positional arguments
kwargs_json='{...}' # Keyword arguments
)
Examples:
# Search customers
execute_method(
model="res.partner",
method="search_read",
args_json='[[["customer_rank", ">", 0]]]',
kwargs_json='{"fields": ["name", "email"], "limit": 20}'
)
# Create sales order
execute_method(
model="sale.order",
method="create",
args_json='[{"partner_id": 8, "order_line": [[0, 0, {"product_id": 5, "product_uom_qty": 1}]]}]'
)
# Confirm order
execute_method(
model="sale.order",
method="action_confirm",
args_json='[[5]]'
)
2. batch_execute - Atomic Transactions
Execute multiple operations atomically. All succeed or all rollback. Use "@N" (1-indexed) inside args_json to reference a previous operation's result.
batch_execute(
operations=[
{"model": "res.partner", "method": "create",
"args_json": '[{"name": "Acme"}]'},
{"model": "sale.order", "method": "create",
"args_json": '[{"partner_id": "@1", "order_line": [[0, 0, {"product_id": 5, "product_uom_qty": 1}]]}]'},
{"model": "sale.order", "method": "action_confirm",
"args_json": '[[@2]]'}
],
atomic=True
)
3. add_cookbook_pattern - Self-Improving Knowledge Base
After ≥4 failed approaches to the same problem, document the working solution so future sessions skip the trial-and-error:
add_cookbook_pattern(
problem="Search products by attribute value",
failed_approaches=[
"Used = on many2many field",
"Tried dotted notation",
"Used wrong model",
"Passed scalar instead of list to in",
],
working_solution='execute_method(model="product.template", method="search_read", args_json=\'[[["product_template_attribute_value_ids", "in", [123]]]]\')',
why_it_works="Many2many fields require in with a list, not =",
key_lesson="For m2m: always 'in' with a list, never '=' with scalar"
)
The ≥4 threshold is enforced — shallow trial-and-error stays out of the cookbook.
📚 Documentation
Discovery resources (read before guessing)
| Resource | Returns |
|---|---|
| odoo://models | All models in this instance |
| odoo://model/{model}/schema | Full schema — fields, types, requireds, relationships |
| odoo://model/{model}/access | Your CRUD permissions on this model |
| odoo://methods/{model} | Methods available on this model |
| odoo://workflows | Business workflows from installed modules |
| odoo://server/info | Odoo version + installed modules |
| odoo://cookbook/patterns | Learned Patterns from past sessions (read after first failure) |
Plus odoo://record/{model}/{id}, odoo://search/{model}/{domain}, odoo://fields/{model}, odoo://model/{model} for one-off reads.
The Cookbook
📖 COOKBOOK.md — 45+ practical examples:
- Searching & filtering
- Creating records
- Updating & deleting
- Working with relationships
- Business workflows
- Batch operations
- Advanced patterns
Start here! The cookbook shows you how to accomplish anything with just 2 tools.
Prompts
User-selectable templates in Claude's menu:
search-customers— Find customers with filterscreate-sales-order— Create sales orders step-by-stepodoo-exploration— Discover your Odoo instance capabilities
Claude Code skills
Eight skills ship in .claude/skills/ and auto-activate on relevant requests when this repo is opened in Claude Code. They wrap the cookbook into focused, trigger-aware guides:
odoo-mcp-searching— domains, operators, name_searchodoo-mcp-efficient-queries— pagination, fields,read_groupodoo-mcp-crud— create/write/unlink, archive vs deleteodoo-mcp-relationships— m2o/o2m/m2m command tuplesodoo-mcp-workflows—action_confirm,action_post,button_validateodoo-mcp-batch— atomic transactions with@Nreferencesodoo-mcp-real-world— HR/CRM/inventory cross-model recipesodoo-mcp-learned-patterns— cookbook read/write workflow
Other MCP clients (Claude Desktop, Cursor) get the same knowledge portably via COOKBOOK.md and the odoo://cookbook/patterns resource.
🎓 Learn with examples
Example 1: Find Employees
execute_method(
model="hr.employee",
method="search_read",
args_json='[[["name", "ilike", "john"]]]',
kwargs_json='{"fields": ["name", "job_id", "department_id"], "limit": 10}'
)
Example 2: Time Off Requests
execute_method(
model="hr.leave",
method="search_read",
args_json='[[
["employee_id", "=", 1],
["date_from", ">=", "2025-01-01"],
["state", "=", "validate"]
]]',
kwargs_json='{"fields": ["employee_id", "date_from", "date_to", "holiday_status_id"]}'
)
Example 3: Create Customer + Order (Atomic)
batch_execute(
operations=[
{
"model": "res.partner",
"method": "create",
"args_json": '[{"name": "Acme Corp", "email": "contact@acme.com"}]'
},
{
"model": "sale.order",
"method": "create",
"args_json": '[{"partner_id": 123, "order_line": [[0, 0, {"product_id": 5}]]}]'
}
],
atomic=True
)
💡 Why This Design Works
✅ Key Benefits
1. Universal Access
- Full Odoo API at your fingertips
- No artificial limitations
- Do anything Odoo can do
2. Simple & Predictable
- Learn 2 tools, use everywhere
- Clear mental model
- Easy to debug and maintain
3. Reliable
- Odoo provides excellent native error messages
- Direct API access means fewer points of failure
- Stable, production-ready implementation
4. Flexible
- Works with several Odoo version (14+)
- Supports all Odoo models and methods
- Extensible through Odoo's native capabilities
🔥 Features
AI Integration
- Claude Desktop / Claude Code / Cursor Ready: Works with any MCP host
- Universal Tools:
execute_method+batch_executereach the entire Odoo API - Self-improving cookbook:
odoo://cookbook/patternsresource +add_cookbook_patterntool grow the institutional knowledge automatically - Smart Limits: Automatic protection against oversized queries (DEFAULT=100, MAX=1000)
- MCP 2025-06-18 spec: Built on FastMCP 2.12+
Multiple Connection Options
- STDIO: Direct integration with Claude Desktop
- SSE: Server-Sent Events for web browsers (port 8009)
- HTTP: Streamable HTTP for API integrations (port 8008)
- Docker: Pre-built containers for all transports
Enterprise Re
Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
117.6kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Agent-Reach
112.7kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
ruflo
82.0k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
headroom
79.9kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
