
Joomla MCP Server
Joomla MCP Server
README
Joomla MCP Server
Table of Contents
Introduction
The Joomla MCP (Model Context Protocol) Server enables AI assistants, such as Claude, to interact with Joomla websites via the Joomla Web Services API. It provides tools to manage articles using an AI assistant.
Features
- Retrieve all articles from a Joomla website
- List all content categories
- Create new articles
- Manage article states (published, unpublished, trashed, or archived)
- Delete articles
- Update articles (requires both introtext and fulltext, with a "Read more" break)
Requirements
- Python 3.11+
- Joomla 4.x or 5.x with the Web Services API plugin enabled
- API Bearer token for authentication
Installation
Installing via Smithery
To install Joomla Articles MCP for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @nasoma/joomla-mcp-server --client claude
Create a Joomla API Token
-
Access User Profile: Log in to the Joomla Administrator interface and navigate to the Users menu, then select Manage.
-
Edit Super User: Find and click on the Super User account (or the desired user) to edit their profile.
-
Generate Token: Go to the Joomla API Token tab, click the Generate button, and copy the displayed token.
Install the Project/Server locally
- Clone the repository:
git clone https://github.com/nasoma/joomla-mcp-sever.git
cd joomla-mcp-server
- Set up a virtual environment and install dependencies using
uv
(a Python dependency manager, see uv documentation). If uv is installed run:
uv sync
Installing on Claude or other AI assistants
Claude Desktop
Add this to your claude_desktop_config.json
:
{
"mcpServers": {
"Joomla Articles MCP": {
"command": "{{PATH_TO_UV}}",
"args": [
"--directory",
"{{PATH_TO_PROJECT}}",
"run",
"main.py"
],
"env": {
"JOOMLA_BASE_URL": "<your_joomla_website_url>",
"BEARER_TOKEN": "<your_joomla_api_token>"
}
}
}
}
Replace {{PATH_TO_UV}}
with the path to uv
(run which uv
to find it) and {{PATH_TO_PROJECT}}
with the project directory path (run pwd
in the repository root).
Available Tools
1. get_joomla_articles()
Retrieves all articles from the Joomla website via its API.
2. get_joomla_categories()
Retrieves all categories from the Joomla website and formats them in a readable list.
3. create_article()
Creates a new article on the Joomla website via its API.
Parameters:
article_text
(required): The content of the article (plain text or HTML)title
(optional): The article title (inferred from content if not provided)category_id
(optional): The category ID for the articleconvert_plain_text
(optional, default: True): Auto-converts plain text to HTMLpublished
(optional, default: True): Publishes the article immediately
4. manage_article_state()
Manages the state of an existing article on the Joomla website via its API.
Parameters:
article_id
(required): The ID of the existing article to check and updatetarget_state
(required): The desired state for the article (1=published, 0=unpublished, 2=archived, -2=trashed)
5. delete_article()
Deletes an article from the Joomla website via its API.
Parameters:
article_id
(required): The ID of the article to delete
6. update_article()
Updates an existing article on the Joomla website via its API. Both introtext
and fulltext
are required to align with Joomla’s article structure (introtext for the teaser, fulltext for the content after a "Read more" break).
Parameters:
article_id
(required): The ID of the article to updatetitle
(optional): New title for the articleintrotext
(required): Introductory text for the article (plain text or HTML)fulltext
(required): Full content for the article (plain text or HTML)metadesc
(optional): Meta description for the article
Security Considerations
- The Joomla API Token has access to your site. Treat it the same way you treat your passwords.
- The server sanitizes HTML content to prevent XSS attacks
- Ensure your Joomla website uses HTTPS to secure API communications.
License
This project is licensed under the MIT License.
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。