daisyui

The DaisyUI MCP Server exposes two main tools to help AI agents interact with DaisyUI component documentation in a token-efficient way. The list_components tool provides a catalog of all available DaisyUI components along with short descriptions, enabling quick discovery of what’s supported. The get_component tool retrieves full documentation for a specific component, including relevant classes, syntax, and practical examples. The component docs are sourced from daisyui.com/llms.txt and stored locally as markdown files, and you can customize or add your own docs to fit your project. To use the server, run the MCP server locally and query it through your editor or integration that supports MCP, choosing the appropriate tool (list_components to browse, get_component to fetch details for a chosen component). The design prioritizes token efficiency by exposing only the necessary context for each query.

Prerequisites:

• Python 3.10+ installed on your system
• Git available to clone the repository
• (Optional) a virtual environment tool like venv

Step 1: Clone the repository

git clone https://github.com/birdseyevue/daisyui-mcp.git
cd daisyui-mcp

Step 2: Create and activate a virtual environment (recommended)

python -m venv venv

# Windows
venv\Scripts\activate

# macOS/Linux
source venv/bin/activate

Step 3: Install dependencies

pip install -r requirements.txt

Step 4: Run the MCP server

python mcp_server.py

Docker alternative:

Build or pull and run the container as described in the repository, for example:

# Pull and run the image
docker run -d -p 8000:8000 --name daisyui-mcp ghcr.io/birdseyevue/daisyui-mcp:latest

If you run via Docker, you can connect either using Stdio (in-editor) or HTTP/SSE as detailed in the README's Docker section.

Tips and common considerations:

• The server exposes two MCP tools: list_components and get_component. Use list_components first to discover available components, then use get_component with the component name to fetch full documentation.
• If you customize component docs in /components, updating via update_components.py is recommended to fetch the latest upstream content while preserving your changes manually. If you want to keep your custom docs, run updates selectively.
• Default port for the Python server is 8000 when running locally. If you use Docker, you can map ports as needed (e.g., -p 8000:8000).
• In editor configurations, ensure the Python interpreter path and the mcp_server.py path reflect your environment (e.g., a virtualenv path or workspace layout).
• The MCP docs are pulled from daisyui.com/llms.txt and stored as markdown files locally; you can edit or add new docs to /components to tailor the repository to your project.