About odoo-lint
odoo-lint is a Rust-native checker for Odoo modules, aimed at Odoo developers and maintainers. It covers multiple module file types and OCA checks, with automatic fixes and an MCP server.
Key features
- 01Check Python, manifests, XML, CSV, PO, and JavaScript
- 02Match pylint-odoo and OCA checks
- 03Apply fixes and preview diffs
- 04Flag leftovers from older Odoo versions
- 05Provide language-server quick fixes
- 06Generate CI output and offer MCP
Use cases
- 01Check Odoo modules before submission
- 02Fix code and translation issues in bulk
- 03Find outdated patterns after an upgrade
- 04Have a coding agent lint files after edits
MCP support
odoo-lint includes an MCP server available through `odl mcp`, allowing MCP clients or coding agents to invoke its checking capabilities. The README does not specify the individual MCP tools.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README
odoo-lint
Blazing fast Rust-native linter for Odoo modules. The command is called odl.
It checks Python files, manifests, XML and CSV data files, translation
(.po) files and JavaScript (OWL components and templates), with the checks
of OCA's pylint-odoo, oca-checks-odoo-module and oca-checks-po: same
names, same messages, same results on their own test suites. It also warns about files that do not belong in a module, such
as packages or database dumps committed by accident.
Linting all of OCA/sale-workflow 18.0, as OCA's pre-commit hook runs pylint-odoo. See benchmarks.
- ⚡ 250x faster than pylint with pylint-odoo, and still 30x faster on a single core
- ⚖️ Drop-in for pylint-odoo and OCA's pre-commit checks: same codes,
names and messages, every check matching on their own test suites
(parity), and
# pylint: disable=comments keep working - 🔧 Automatic fixes, safe and unsafe as in Ruff, for Python, manifests,
XML and
.pofiles (odl check --fix,--diffto preview) - ⬆️ Upgrade checks: leftovers of older Odoo versions, such as Bootstrap 4 classes that do nothing since Odoo 15.0, with fixes
- 🌍 Translation checks that predict Odoo's loader: why Odoo logs "malformed po file", and why a translation does not show up
- 📦 Installable with pip, uv or pipx: one binary in prebuilt wheels for Linux, macOS and Windows, without Python dependencies
- 🛠️ Configured in
pyproject.toml, including company-specific author rules - 📋 Output for CI: GitHub annotations, GitLab Code Quality, SARIF and JSON
- ✍️ In your editor: a language server (
odl server) with quick fixes, and extensions for VS Code and Zed - 🤖 For AI coding agents: an MCP server, and plugins for Claude Code, DeepSeek Harness and OpenCode that lint every file the agent edits
- 🪝 Git hooks: pre-commit and hk
Installation
uv tool install odoo-linter # or: pipx install odoo-linter
cargo install odoo-lint # from crates.io
The PyPI package is called odoo-linter, because odoo-lint is too similar
to an existing PyPI project. On crates.io it is odoo-lint.
Quick start
odl check path/to/addons # lint a directory
odl rule # list all rules
odl rule C8101 # explain a rule
odl check --fix # apply the safe fixes (--diff to preview)
Rules ported from pylint-odoo keep its codes, names and messages, and
# pylint: disable= comments keep working. Company-specific rules are
configured in your pyproject.toml:
[tool.odoo-lint]
target-version = "17.0"
[tool.odoo-lint.rules.manifest-required-author.mapping]
"acme_*" = "Acme Corp"
For AI coding agents there is an MCP server (odl mcp) and a post-edit hook,
packaged as plugins for Claude Code (/plugin marketplace add bosd/odoo-lint),
DeepSeek Harness and OpenCode.
New to odoo-lint? Start with Getting started; coming from OCA's pre-commit hooks, read Migrating from OCA's tooling. See the documentation for all commands, configuration options and rules.
Development
uv sync --all-groups # builds odl into .venv via maturin
uv run nox # pre-commit, fmt, clippy, cargo tests, CLI tests, docs
See CONTRIBUTING.md and the development guide.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.