odoo-devkit-mcp
A token-efficient Model Context Protocol (MCP) server for Odoo 17/18/19 development. Gives AI assistants (Claude, Cursor, etc.) fast, structured access to your Odoo codebase — module discovery, model analysis, view navigation, XML validation, and code scaffolding — without loading entire files into context.
Documentation
odoo-devkit
Token-efficient MCP server for Odoo 17/18/19 development.
`odoo-devkit` gives AI assistants structured access to your Odoo codebase for discovery, search, model/view analysis, XML checks, and scaffolding patches.
Highlights
- Fast module and code discovery across addons roots
- Model surface inspection (fields, methods, views, actions, menus, security)
- View inheritance navigation (`find_view_chain`, inherited views, field usage)
- Patch scaffolding for common Odoo tasks (models, views, actions, menus, security, wizards, reports)
- XML validation for view structure and field references
- Web dashboard that starts with the MCP server (Serena-style flow)
- Layered config: global + project override
Requirements
- Python 3.10+
- `uv` (recommended)
- `ripgrep` (`rg`) for fast search
Install examples:
# Ubuntu/Debian
sudo apt install ripgrep
# macOS
brew install ripgrepInstallation
git clone https://github.com/VatsalChauhan36/odoo-devkit.git
cd odoo-devkit
uv sync
uv run odoo-devkit --helpConfiguration
You can configure roots and defaults via environment variables, global config, and project config.
Config files (Serena-like)
- Global config: `~/.odoo-devkit/config.json`
- Project config: `/.odoo-devkit/project.json`
Effective config resolution:
1. Global config
2. Project override (if project root is detected)
Project root detection order:
1. `ODOO_MCP_PROJECT_ROOT`
2. nearest ancestor containing `.odoo-devkit/project.json`
3. nearest ancestor containing `.git`
Root paths priority
Addons roots are resolved with this priority:
1. `ODOO_MCP_ROOTS`
2. saved config (`roots` in global/project config)
Typical roots order:
1. Custom addons
2. Odoo server addons
3. Odoo base addons
Example:
export ODOO_MCP_ROOTS="/path/to/custom-addons:/path/to/odoo/addons:/path/to/odoo/odoo/addons"Windows PowerShell:
$env:ODOO_MCP_ROOTS = "C:\custom-addons;C:\odoo\addons;C:\odoo\odoo\addons"Optional docs search
To enable `search_odoo_docs`, set:
export ODOO_MCP_DOCS_PATH="/path/to/odoo/documentation"Dashboard flow
The dashboard starts with the MCP server by default.
Saved settings include:
- `open_browser`
- `enable_dashboard`
- `dashboard_host`
CLI overrides:
uv run odoo-devkit --no-dashboard
uv run odoo-devkit --dashboard-host 0.0.0.0
uv run odoo-devkit --no-open-browser
uv run odoo-devkit --project-root /path/to/projectIf browser auto-open is skipped (for example, headless Linux), use the MCP tool `open_dashboard`.
Screenshots
Then this preview will render automatically:

Deprecated CLI flags
These flags are intentionally kept for backward compatibility and are ignored:
- `--roots`
- `--config`
They emit warnings and do not affect runtime behavior.
MCP client setup
Claude Code
{
"mcpServers": {
"odoo-devkit": {
"type": "stdio",
"command": "uv",
"args": [
"run",
"--directory", "/path/to/odoo-devkit",
"odoo-devkit",
"--project-root", "/path/to/your/project"
]
}
}
}Cursor
{
"mcpServers": {
"odoo-devkit": {
"command": "uv",
"args": [
"run",
"--directory", "/path/to/odoo-devkit",
"odoo-devkit",
"--project-root", "/path/to/your/project"
]
}
}
}Tool overview
Discovery & search
- `list_modules`
- `list_custom_modules`
- `get_module_manifest`
- `get_module_structure`
- `glob_odoo_files`
- `read_file_lines`
- `search_odoo_code`
- `search_odoo_docs`
Model and view analysis
- `inspect_model_surface`
- `find_model_definition`
- `get_model_fields`
- `find_method_definition`
- `find_view_definition`
- `find_inherited_views`
- `find_view_by_model`
- `find_view_chain`
- `find_field_in_views`
- `validate_view_xml`
Actions, menus, and security
- `find_action_by_model`
- `find_menu_hierarchy`
- `find_security_access_for_model`
Scaffolding (patch generation)
- `scaffold_model_patch`
- `scaffold_inherit_model_patch`
- `scaffold_view_inherit_patch`
- `scaffold_views_patch`
- `scaffold_action_patch`
- `scaffold_menu_patch`
- `scaffold_security_access_patch`
- `scaffold_record_rule_patch`
- `scaffold_wizard_patch`
- `scaffold_report_patch`
- `manifest_update_patch`
- `init_update_patch`
Runtime helpers
- `run_module_upgrade`
- `execute_rpc`
- `check_rpc_connection`
- `open_dashboard`
Recommended usage pattern
1. Start with `list_custom_modules` and `get_module_structure`
2. Narrow with `search_odoo_code` (`module_filter`, `glob`)
3. Use `inspect_model_surface` for model-level orientation
4. Use view tools for inheritance-aware XML navigation
5. Use scaffold tools to generate safe patches before applying
Troubleshooting
`No valid roots found`
Set `ODOO_MCP_ROOTS` or save roots in dashboard config (global/project).
`rg: command not found`
Install ripgrep (`ripgrep` package).
Dashboard did not open
- Check `open_browser` setting
- Use `--open-browser` to force when supported
- On headless Linux, open URL manually or call `open_dashboard`
MCP client cannot connect
- Verify path in `--directory`
- Re-run `uv sync`
- Run `uv run odoo-devkit --help` locally
License
MIT. See LICENSE.
Frequently asked questions
What is odoo-devkit-mcp?
odoo-devkit-mcp is A token-efficient Model Context Protocol (MCP) server for Odoo 17/18/19 development. Gives AI assistants (Claude, Cursor, etc.) fast, structured access to your Odoo codebase — module discovery, model analysis, view navigation, XML validation, and code scaffolding — without loading entire files into context.
How do I install odoo-devkit-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is odoo-devkit-mcp open source?
Yes — it is hosted on GitHub at https://github.com/VatsalChauhan36/odoo-devkit-mcp and has 4 stars.
Related MCP tools
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
Automate browser based workflows with AI
Hindsight: Agent Memory That Learns
A privacy-first app that strips AI watermarks from content you own.
Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
The power of Claude Code / GeminiCLI / CodexCLI + [Gemini / OpenAI / OpenRouter / Azure / Grok / Ollama / Custom Model / All Of The Above] working as one.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP