cellar-wrapper
A community-maintained MCP server that simplifies access to EU legal and legislative data from the CELLAR service, supporting lookups, metadata retrieval, relation checks, and monitoring.
README
cellar-wrapper
cellar-wrapper helps you get legal and legislative data from the EU Publications Office CELLAR service in a simpler, more practical way.
You can use it:
- as a Python library,
- from the command line,
- and, if needed, as an MCP server for AI/tool integrations.
This project is community-maintained and unofficial. It is not affiliated with, endorsed by, or operated by the European Union or the Publications Office of the European Union.
Important status
This is an early-stage project.
The Python API, CLI commands, MCP interface, and response shapes may change. Do not treat the current interface as final or stable.
What it is for
The project is meant for people who need to:
- look up an EU act by CELEX number,
- fetch basic metadata for an act,
- check relations such as amendments, repeals, citations, and related case law,
- monitor whether something new appeared after a given date,
- download text or legislative summaries,
- search acts by title, EuroVoc, or subject matter.
In short: it is a practical access layer over CELLAR for legal research, compliance work, and structured data collection.
Current state and limitations
This is a hobby project, not an official product.
Before using it seriously, keep in mind:
- the project is still evolving, so commands and outputs may change,
- some data comes directly from CELLAR and can be incomplete, sparse, or inconsistent,
- not every legal question maps neatly to one command,
- MCP support exists, but the main practical use today is still Python and the CLI,
Installation
Requires Python 3.11 or newer.
If you want the published package, install the library and CLI with:
pip install cellar-wrapper
If you want the current repository version instead, install directly from GitHub:
pip install "git+https://github.com/Kymylyy/cellar-wrapper.git"
For library, CLI, and MCP support:
pip install "cellar-wrapper[mcp]"
Or from GitHub with the MCP extra:
pip install "cellar-wrapper[mcp] @ git+https://github.com/Kymylyy/cellar-wrapper.git"
Quick use from the command line
For most people, the CLI is the easiest way to start.
Check that the CLI is installed:
cellar --version
Resolve a CELEX number:
cellar lookup resolve-celex --celex 32022R2554
Get metadata for one act:
cellar lookup get-act --celex 32022R2554
Check amendments:
cellar relations get-amendments --celex 32022R2554 --limit 50
Check what is new since a date:
cellar monitoring new-citations --celex 32022R2554 --since 2025-01-01
The CLI returns JSON. On success:
{"ok": true, "data": {...}}
On error:
{"ok": false, "error": {"type": "CellarValidationError", "message": "...", "details": {...}}}
Quick use from Python
from cellar_wrapper import CellarClient
with CellarClient() as client:
act = client.get_act("32022R2554")
print(act.celex, act.title)
Help, bugs, and security
If something does not work, or if you want to ask how to use the project:
- Support and usage questions: SUPPORT.md
- Bug reports and feature requests: https://github.com/Kymylyy/cellar-wrapper/issues
- Contributing: CONTRIBUTING.md
- Security reporting: SECURITY.md
Quick use as MCP
If you want to connect the project to an assistant or another tool through MCP, install the MCP extra:
pip install "cellar-wrapper[mcp]"
Run the server:
python -m cellar_wrapper.mcp_server
Check the version:
python -m cellar_wrapper.mcp_server --version
Environment variables supported by the MCP server:
CELLAR_MCP_BASE_URL_SPARQLCELLAR_MCP_BASE_URL_RESOURCECELLAR_MCP_USER_AGENTCELLAR_MCP_RETRIESCELLAR_MCP_TIMEOUT_CONNECTCELLAR_MCP_TIMEOUT_READCELLAR_MCP_TIMEOUT_WRITECELLAR_MCP_TIMEOUT_POOL
Where to look next
- Docs index: https://github.com/Kymylyy/cellar-wrapper/blob/main/docs/README.md
- Command guide: https://github.com/Kymylyy/cellar-wrapper/blob/main/docs/COMMAND_GUIDE.md
- Curated examples: https://github.com/Kymylyy/cellar-wrapper/blob/main/docs/CONTRACT_EXAMPLES.md
- Contract reference: https://github.com/Kymylyy/cellar-wrapper/blob/main/docs/CONTRACT_REFERENCE.md
For contributors
Main local checks:
ruff check
mypy
pytest -m "not live"
python -m build --sdist --wheel
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。