🚀 Fast, local, offline-first MCP server for Shopify Liquid documentation with 198 comprehensive docs including all tags, filters, and objects.
A specialized Model Context Protocol (MCP) server that provides instant access to complete Shopify Liquid documentation for AI assistants like Claude, Cursor, and other MCP-compatible tools.
While Shopify provides an official MCP server that covers all Shopify APIs, this server is specifically optimized for Liquid template development:
| Feature | Official Shopify MCP | This (Liquid MCP) |
|---|---|---|
| Focus | All Shopify APIs (Admin, Storefront, Functions, etc.) | Liquid templating only |
| Storage | Remote (queries shopify.dev) | Local (no network needed) |
| Speed | Network-dependent | Instant (<1ms) |
| Offline | ❌ Requires internet | ✅ Fully offline |
| Customization | Fixed documentation | Add your own docs |
| Coverage | General API docs | 198 Liquid-specific docs |
| Runtime | Node.js | Python |
Use Official Shopify MCP for:
- Admin API/GraphQL development
- App development with Shopify APIs
- Functions and POS extensions
- Real-time schema validation
Use This Liquid MCP for:
- Theme development with Liquid
- Liquid template debugging
- Offline development
- Faster Liquid reference lookups
- Project-specific Liquid snippets
Use Both! They complement each other perfectly - official MCP for APIs, this for Liquid.
- 🚀 Lightning Fast - Local SQLite FTS5 search (<1ms response time)
- 📚 Complete Coverage - 30 tags, 101 filters, 67 objects (198 total docs)
- 🔌 Offline First - No network requests, works anywhere
- 🐳 Docker Ready - One command deployment
- 🎨 IDE Integration - Works with VS Code, Cursor, Claude Desktop, Zed
- 🔧 Customizable - Add your own project-specific Liquid docs
- 📖 Rich Documentation - Full syntax, parameters, examples for everything
- 🔍 Smart Search - Full-text search with snippet highlighting
pip install git+https://github.com/florinel-chis/shopify-liquid-mcp.git# Pull the image
docker pull florinel-chis/shopify-liquid-mcp:latest
# Or build locally
docker build -t shopify-liquid-mcp .
docker run -it shopify-liquid-mcpdocker-compose up -dgit clone https://github.com/florinel-chis/shopify-liquid-mcp.git
cd shopify-liquid-mcp
pip install -e .Create or edit .vscode/settings.json:
{
"mcp.servers": {
"shopify-liquid": {
"type": "docker",
"image": "shopify-liquid-mcp:latest",
"transport": "stdio"
}
}
}Create .mcp.json in your project root:
{
"mcpServers": {
"shopify-liquid": {
"type": "stdio",
"command": "shopify-liquid-mcp"
}
}
}Or add to VS Code settings:
{
"mcp.servers": {
"shopify-liquid": {
"command": "shopify-liquid-mcp",
"args": [],
"transport": "stdio"
}
}
}Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"shopify-liquid": {
"type": "stdio",
"command": "docker",
"args": ["run", "-i", "--rm", "shopify-liquid-mcp:latest"]
}
}
}{
"mcpServers": {
"shopify-liquid": {
"command": "shopify-liquid-mcp"
}
}
}Add to Cursor's MCP configuration (Settings > Features > MCP Servers):
{
"mcpServers": {
"shopify-liquid": {
"command": "shopify-liquid-mcp"
}
}
}Or with Docker:
{
"mcpServers": {
"shopify-liquid": {
"command": "docker",
"args": ["run", "-i", "--rm", "shopify-liquid-mcp:latest"]
}
}
}Create .zed/settings.json:
{
"context_servers": {
"shopify-liquid": {
"command": "shopify-liquid-mcp"
}
}
}Edit ~/.continue/config.json:
{
"mcpServers": [
{
"name": "shopify-liquid",
"command": "shopify-liquid-mcp"
}
]
}Once configured, your AI assistant has access to 7 specialized tools:
Full-text search across all Shopify Liquid documentation.
Example:
AI: I'll search for "for loop" in the documentation...
Get complete documentation for a specific tag.
Example:
AI: Let me get the documentation for the 'for' tag...
Get complete documentation for a specific filter.
Example:
AI: I'll look up the 'upcase' filter documentation...
Get complete documentation for a specific object.
Example:
AI: Let me check the 'product' object properties...
List all 30 available tags organized by category.
List all 101 available filters organized by category.
List all 67 available objects organized by category.
Try asking your AI assistant:
Getting Documentation:
- "Show me documentation for the for loop tag"
- "How does the money filter work in Liquid?"
- "What properties are available on the product object?"
Searching:
- "Search for documentation about cart functionality"
- "Find all filters related to dates"
- "What objects can I use for collections?"
Listing:
- "List all Shopify Liquid tags"
- "Show me all available string filters"
- "What iteration tags are available?"
Building:
- "Help me display products in a grid"
- "How do I format dates in Liquid?"
- "Show me how to work with the cart"
-
30 Liquid Tags
- Control flow:
if,unless,case,else - Iteration:
for,break,continue,cycle,tablerow,paginate - Variables:
assign,capture,increment,decrement,echo - Theme:
layout,section,sections,render,include,content_for - HTML/Assets:
form,style,stylesheet,javascript - Syntax:
comment,raw,liquid,doc
- Control flow:
-
101 Liquid Filters
- String (24):
upcase,downcase,capitalize,append,prepend, etc. - Array (13):
join,sort,reverse,map,where,size, etc. - Math (11):
plus,minus,times,divided_by,modulo, etc. - Money (4):
money,money_with_currency, etc. - Image (5):
image_url,image_tag, etc. - Color (9):
color_darken,color_lighten, etc. - And more...
- String (24):
-
67 Liquid Objects
- Store:
shop,settings,theme,brand - Products:
product,variant,collection,collections - Cart:
cart,line_item,checkout - Customer:
customer,address,company - Content:
page,blog,article,comment - Media:
image,video,media,model - And more...
- Store:
# Build the image
docker build -t shopify-liquid-mcp .
# Run the server
docker run -it --rm shopify-liquid-mcp
# Run with persistent data
docker run -it --rm -v mcp-data:/data shopify-liquid-mcp# Start the server
docker-compose up -d
# View logs
docker-compose logs -f
# Stop the server
docker-compose down
# Shell access
docker-compose exec shopify-liquid-mcp bashAdd your project-specific Liquid docs:
# Create custom docs directory
mkdir custom-docs
# Add your .md files
cp my-snippets.md custom-docs/
# Mount in docker-compose.yml
volumes:
- ./custom-docs:/docs/custom:ro# Run all tests
pytest
# Run specific test
pytest tests/test_ingest.py
# Run with coverage
pytest --cov=shopify_liquid_mcp# Test the server
python test_server.py
# Test indexing
python -m shopify_liquid_mcp.ingest
# Check database
sqlite3 ~/.mcp/shopify-liquid-docs/database.db "SELECT COUNT(*) FROM liquid_docs;"# Force reindex
python -m shopify_liquid_mcp.ingest --force
# Custom database location
SHOPIFY_LIQUID_DB_PATH=/custom/path/db.sqlite python -m shopify_liquid_mcp.ingestContributions welcome! Please read our Contributing Guide.
# Fork and clone
git clone https://github.com/florinel-chis/shopify-liquid-mcp.git
cd shopify-liquid-mcp
# Create virtual environment
python -m venv venv
source venv/bin/activate # or `venv\Scripts\activate` on Windows
# Install for development
pip install -e ".[dev]"
# Run tests
pytest
# Make changes and submit PRMIT License - see LICENSE file for details.
- Inspired by Gemini API Docs MCP
- Built with FastMCP
- Documentation from Shopify Liquid API
- MCP Protocol by Anthropic
- Official Shopify Dev MCP - For Shopify API development
- FastMCP - MCP server framework
- Model Context Protocol - MCP specification
- 🐛 Issues
- 💬 Discussions
Made with ❤️ for the Shopify theme development community