Quick facts
- Price
- $299.00 USD
- Odoo versions
- 17, 18, 19
- Setup time
- 30 min
- License
- One-time perpetual
- Languages
- 11 languages
One-time Purchase
Advanced MCP Server for Odoo
A build-to-order integration that exposes your live Odoo data and actions to AI assistants through the Model Context Protocol, so teams can query records, generate reports, and trigger workflows in natural language — with Odoo's own access rights and record rules enforced on every call.
$299.00USD
No payment now. This sends a quote request to our team — we'll follow up by email with pricing and next steps.
Key Features
MCP server implementing the Model Context Protocol with typed tool definitions (query, read-document, run-report, invoke-action) that AI assistants discover automatically
Every session authenticates as a real Odoo user via the XML-RPC/JSON-RPC API, so ir.model.access.csv grants and record rules are enforced on all reads and writes
Companion Odoo module with a proper __manifest__.py, models/ extending models.Model, and a declarative registry of which models and fields are exposed or masked
Natural-language queries compiled into validated ORM search_read calls with domain filters, field projection, and pagination — never raw SQL
Multi-company aware: company_id boundaries and record rules honored so the assistant never crosses a company the user cannot see
Write guardrails: allow-listed models/fields, per-tool rate limits, batch-size caps, and mandatory confirmation gates on financial or state-changing actions
Whitelisted QWeb report execution so the assistant can generate the exact invoice, SO, or picking PDFs/XLSX your users already use
Whitelisted ir.actions.server / automated-action triggering for scoped workflow automation (e.g. confirm a draft, post a reminder, tag a lead)
Full audit trail model capturing user, tool, arguments, and affected record IDs for every AI-initiated call
Field-level masking layer to redact sensitive columns (salaries, cost prices, personal data) even from users who can technically read the model
Compute/@api.depends helper fields where the assistant needs derived values without recomputing logic client-side
Built and regression-tested for Odoo 17.0, 18.0, and 19.0 on both Community and Enterprise, with edition-gated tools
Deployment options for self-hosted, Odoo.sh, or on-prem, with credentials handled via environment/secret store rather than hardcoded keys
Optional streaming of large result sets and cursor-based pagination to keep AI context windows within limits
About this Product
Most teams that run Odoo already have the data they need locked inside the ORM — sales orders, stock moves, invoices, project tasks, CRM leads — but getting an answer still means clicking through menus, building a filtered list view, or exporting to a spreadsheet. The moment you want an AI assistant to help ("what changed on this customer's overdue invoices this week, and draft a follow-up"), Odoo native runs out of road: there is no supported bridge that lets a modern AI client read live records, respect a given user's ir.model.access.csv and record rules, and act on the result without someone hand-writing XML-RPC glue for every request. Bolting a raw API key onto a chatbot is worse — it ignores multi-company boundaries, leaks fields the user should never see, and hands an unpredictable model unbounded write access to production.
Why Choose Advanced MCP Server for Odoo
MCP server implementing the Model Context Protocol with typed tool definitions (query, read-document, run-report, invoke-action) that AI assistants discover automatically
Every session authenticates as a real Odoo user via the XML-RPC/JSON-RPC API, so ir.model.access.csv grants and record rules are enforced on all reads and writes
Companion Odoo module with a proper __manifest__.py, models/ extending models.Model, and a declarative registry of which models and fields are exposed or masked
Natural-language queries compiled into validated ORM search_read calls with domain filters, field projection, and pagination — never raw SQL
Multi-company aware: company_id boundaries and record rules honored so the assistant never crosses a company the user cannot see
Write guardrails: allow-listed models/fields, per-tool rate limits, batch-size caps, and mandatory confirmation gates on financial or state-changing actions
The Advanced MCP Server is a made-to-order Odoo integration that ECOSIRE builds, installs, and supports for you. It implements the Model Context Protocol as a thin, security-first server that sits in front of your Odoo instance and speaks to AI assistants as a set of well-defined tools: query records, read a specific document, run a saved report, or invoke a whitelisted automated action. Under the hood each session authenticates as a real Odoo user and calls the ORM through the XML-RPC/JSON-RPC API — so every read and write is filtered by that user's groups, ir.model.access.csv grants, and record rules exactly as if they were logged into the web client. There is no god-mode key: if a sales rep can't see another team's pipeline in Odoo, the assistant can't either.
Technically, we ship the MCP layer plus a companion Odoo module (a proper __manifest__.py, models/ extending models.Model, and security definitions) that declares the tool surface: which models are queryable, which fields are exposed or masked, which QWeb reports and ir.actions.server automated actions the assistant may trigger, and hard guardrails on writes (allow-listed models/fields, per-tool rate limits, and confirmation gates on anything financial). Query tools translate a natural-language ask into a validated search_read with domain, field projection, and pagination — never raw SQL — while report tools stream the same PDFs/spreadsheets your users already trust. Every call is logged to an audit trail (user, tool, arguments, affected records) so you can see precisely what the AI did. It runs against Community or Enterprise and is built and tested for Odoo 17.0, 18.0, and 19.0; Enterprise-only surfaces (for example studio-built models or accounting features) are wired only when your edition actually has them.
Delivery is build-to-order. After a short scoping call we confirm the exact models, reports, and actions to expose, agree the security matrix, then build against a copy of your database, hand you the module and MCP server on a staging environment for UAT, and go live only once you sign off — typically 2–4 weeks from confirmed scope. You receive the full source, a git repo handover, and a support window. Pricing starts from $299 (indicative, single-company base scope); broader tool surfaces, multi-company record-rule design, additional AI-client integrations, and extra automated-action workflows increase the quoted scope.
What you get
- Installable source code for your version of the MCP server plus the companion Odoo module (technical name confirmed at scoping)
- Installation and configuration on your environment (self-hosted, Odoo.sh, or on-prem), including AI-client connection setup
- Technical documentation: architecture, the exposed tool/model/field matrix, the security model, and the audit-log schema
- End-user guide plus a live training session for the team and administrators
- UAT on a staging environment against a copy of your database, with a documented rollback plan before go-live
- Post-go-live support window for defect fixes and configuration questions
- Git repository handover with full commit history and branch/tag conventions
- Security matrix sign-off document mapping every exposed model, field, report, and automated action to the Odoo groups allowed to reach it
Who this is for
Operations & finance managers running Odoo daily
They want instant answers about orders, invoices, inventory, and cash without building a new list view or export each time — and need confidence the AI only sees and touches what their own Odoo login allows.
IT leads and Odoo administrators
They need a supported, auditable bridge between AI tools and production Odoo that enforces ir.model.access.csv and record rules, avoids god-mode API keys, and can be reviewed, versioned, and rolled back safely.
Sales and CRM team leaders
They want to ask about their pipeline, draft follow-ups, and trigger scoped actions in natural language while multi-company and team-level record rules keep each rep inside their own data.
AI/automation champions building assistant workflows
They need clean MCP tools over live Odoo — query, report, and whitelisted actions — instead of hand-writing XML-RPC glue for every integration, with rate limits and confirmation gates baked in.
Installation overview
Download from ECOSIRE
Buy the license on ecosire.com and download the Advanced MCP Server for Odoo module ZIP from your account dashboard.
Upload to Odoo
Extract the ZIP into your Odoo custom addons folder on the server (or upload via Apps > Install from file on Odoo.sh / runbot).
Install the module
Activate Developer Mode, open Apps, click Update Apps List, search for Advanced MCP Server for Odoo, and press Install.
Configure settings
Open the new menu, paste your ECOSIRE license key, connect any external credentials (Shopify, Amazon, Stripe, etc.), and save.
Verify & go live
Run the built-in connection test, sync your first 10 records, and schedule the recurring cron. Contact support if anything fails.
How Advanced MCP Server for Odoo Compares
| Criterion | ECOSIRE | Custom Build | Competitor | Odoo Native |
|---|---|---|---|---|
| How you get it | Build-to-order: scoped, built, installed, and supported by ECOSIRE | Your team designs and codes the whole MCP + module in-house | Instant apps.odoo.com download you configure yourself | |
| Security enforcement | Real-user auth; ir.model.access.csv + record rules enforced on every call | Depends entirely on your own implementation discipline | Often a shared API key; record-rule handling varies | |
| AI write guardrails | Allow-listed models/fields, rate limits, confirmation gates, audit log | Whatever you have time to build and test | Usually generic; limited action-level gating | |
| Version coverage | Built and tested for 17.0/18.0/19.0, Community + Enterprise | You maintain compatibility across versions yourself | Pinned to specific versions; upgrades on the vendor's timeline | |
| Fit to your data | Tool/field matrix tailored to your models, reports, and workflows | Fully bespoke, at full build cost and timeline | Generic surface; you adapt your process to the module | |
| Reports & actions | Whitelisted QWeb reports + scoped ir.actions.server triggers | You wire each report and action integration by hand | Limited or fixed set of supported reports/actions | |
| Ownership & handover | Full source, git repo handover, technical docs, training | You own it because you built it — plus the maintenance burden | Vendor keeps the source; you get a black-box install | |
| Total cost profile | From $299 indicative; fixed quote after scoping, support included | Highest — senior dev time to build, secure, and maintain | Low sticker price; hidden cost in fit, security, and support |
Frequently Asked Questions about Advanced MCP Server for Odoo
How long does delivery take?
This is a build-to-order product, not an instant download. After a scoping call to confirm the models, reports, actions, and security matrix, typical delivery is 2–4 weeks from confirmed scope, including a UAT window on staging before go-live. Larger tool surfaces or multi-company designs can extend that, which we tell you up front.
How does pricing and scoping work?
Pricing starts from $299 (indicative, single-company base scope). We hold a short scoping call, confirm exactly which models/fields/reports/actions to expose and your edition, then give you a fixed written quote before any build starts. Drivers like multi-company record-rule design, extra AI-client integrations, and additional automated-action workflows increase the quoted scope.
Is my Odoo data safe from the AI seeing or changing the wrong thing?
Yes. Every session authenticates as a real Odoo user, so all reads and writes are filtered by that user's groups, ir.model.access.csv grants, and record rules — the assistant can never see or touch data the logged-in user couldn't. Writes are further restricted to allow-listed models/fields with rate limits and confirmation gates, and every call is recorded in an audit log.
Which Odoo versions and editions are supported?
It is built and tested for Odoo 17.0, 18.0, and 19.0 on both Community and Enterprise. Enterprise-only surfaces are wired only when your edition actually provides them; we confirm your exact version and edition during scoping so the delivered build matches your instance.
What about support and updates after go-live?
Every engagement includes a post-go-live support window for defect fixes and configuration questions, plus a git repo handover so your team owns the source. Beyond the included window we offer ongoing support and can adapt the tool surface, add integrations, or migrate the build to a newer Odoo version as your needs change.
Do I have to give an AI assistant a full admin API key?
No — that is exactly the anti-pattern this avoids. There is no god-mode key. The MCP server brokers access through a scoped Odoo user and a declared tool surface, so the assistant is bounded by Odoo's own security model rather than an unrestricted API token.
Which AI assistants can connect to it?
Any AI client that speaks the Model Context Protocol can connect and auto-discover the tools we expose. We confirm your target AI client(s) during scoping and configure the connection during installation; supporting additional clients is part of the quoted scope.
Related Modules
Request a quote
Advanced MCP Server for Odoo
A build-to-order integration that exposes your live Odoo data and actions to AI assistants through the Model Context Protocol, so teams can query records, generate reports, and trigger workflows in natural language — with Odoo's own access rights and record rules enforced on every call.
- MCP server implementing the Model Context Protocol with typed tool definitions (query, read-document, run-report, invoke-action) that AI assistants discover automatically
- Every session authenticates as a real Odoo user via the XML-RPC/JSON-RPC API, so ir.model.access.csv grants and record rules are enforced on all reads and writes
- Companion Odoo module with a proper __manifest__.py, models/ extending models.Model, and a declarative registry of which models and fields are exposed or masked
- Natural-language queries compiled into validated ORM search_read calls with domain filters, field projection, and pagination — never raw SQL