PrestaShop MCP Server
Provides instant, offline access to comprehensive PrestaShop development documentation including 647+ hooks, module guides, component architecture, APIs, and theme development resources for AI assistants.
README
PrestaShop MCP Server
🚀 Fast, local, offline-first MCP server for PrestaShop documentation with 1,095+ indexed documents including hooks, guides, components, and APIs.
A specialized Model Context Protocol (MCP) server that provides instant access to comprehensive PrestaShop development documentation for AI assistants like Claude, Cursor, and other MCP-compatible tools.
🎯 Why This MCP Server?
PrestaShop is a comprehensive e-commerce platform with 647+ hooks, complex component architecture, and extensive APIs. This MCP server provides instant, offline access to 1,095+ indexed documents:
- 647+ PrestaShop Hooks - All display and action hooks with complete metadata
- Module Development - Complete guides for creating and testing modules
- Component Architecture - Forms, Grids, Services, CQRS patterns (119 components)
- Theme Development - Template system, Smarty/Twig documentation
- API Documentation - Admin API, webservice, and domain references (96 APIs)
- Guides & Tutorials - Installation, deployment, contribution guides
vs. Official PrestaShop Documentation
| Feature | Official Docs Website | This (PrestaShop MCP) |
|---|---|---|
| Access | Online only | Local + Offline |
| Speed | Network-dependent | Instant (<50ms) |
| Search | Web search | SQLite FTS5 full-text |
| AI Integration | Manual lookup | Direct MCP integration |
| Customization | Fixed | Add your own docs |
✨ Features
- 🚀 Lightning Fast - Local SQLite FTS5 search (<50ms response time)
- 📚 Comprehensive Coverage - 1,095+ documents: hooks, guides, components, APIs
- 🔌 Offline First - No network requests, works anywhere
- 🐳 Docker Ready - Easy deployment with intelligent wrapper script
- 🎨 IDE Integration - Works with Claude Desktop, Claude Code, VS Code, Cursor
- 🔧 Customizable - Add your own project-specific PrestaShop docs
- 📖 Rich Documentation - Full syntax, parameters, examples, and metadata
- 🔍 Smart Search - Full-text search with type, category, and origin filtering
- 🌐 Multiple Transports - STDIO (local), HTTP, or SSE (remote deployment)
📦 Installation
Option 1: Using Docker (Easiest - Recommended)
git clone https://github.com/florinel-chis/prestashop-mcp.git
cd prestashop-mcp
# Using the wrapper script (best for Claude Desktop/Code)
chmod +x run-docker-mcp.sh
./run-docker-mcp.sh
# Or using Docker Compose (for HTTP/remote deployment)
docker compose up -d
The wrapper script automatically:
- Builds the Docker image if missing
- Handles documentation mounting or auto-fetching
- Configures STDIO transport for MCP clients
- No TTY issues!
See DOCKER_USAGE.md for complete Docker setup guide.
Option 2: Using pip (Recommended for Python)
pip install git+https://github.com/florinel-chis/prestashop-mcp.git
Option 3: From Source
git clone https://github.com/florinel-chis/prestashop-mcp.git
cd prestashop-mcp
pip install -e .
📚 Documentation Setup
The MCP server requires access to the PrestaShop official documentation repository for indexing.
Docker Setup (Automatic)
Docker automatically handles documentation:
- On first run, it clones the PrestaShop docs repository
- Documentation is baked into the Docker image during build
- No manual setup needed!
Local Python Setup (Manual)
For local installations, clone the documentation repository:
# Clone PrestaShop docs as a sibling directory to prestashop-mcp
cd /path/to/your/projects/
git clone https://github.com/PrestaShop/docs.git prestashop-docs
# Directory structure should look like:
# /path/to/your/projects/
# ├── prestashop-mcp/ (this repository)
# └── prestashop-docs/ (PrestaShop official docs)
# The MCP server will automatically find it
Or set a custom location:
# Set environment variable
export PRESTASHOP_DOCS_PATH=/path/to/prestashop-docs
# Then run the server
prestashop-mcp
Documentation Path Priority
The server looks for documentation in this order:
- Environment variable:
PRESTASHOP_DOCS_PATH(if set) - Sibling directory:
../prestashop-docs(relative to MCP server) - Inside project:
./prestashop-docs(for Docker builds)
First Run Indexing
On first run, the server will:
- Locate the PrestaShop documentation repository
- Parse and index 1,095+ markdown files
- Extract metadata (hooks, components, APIs, guides)
- Build SQLite FTS5 search index with Porter stemming
- Store in
~/.mcp/prestashop-docs/database.db
Indexing takes ~30-60 seconds on first run, then uses the cached database for instant startup.
Force reindexing:
# Local installation
python -m prestashop_mcp.ingest_v2 --force
# Or delete the database
rm ~/.mcp/prestashop-docs/database.db
🎮 Usage
With Claude Desktop
Standard Installation:
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"prestashop": {
"command": "prestashop-mcp"
}
}
}
Docker Installation (using wrapper script):
{
"mcpServers": {
"prestashop": {
"command": "/absolute/path/to/prestashop-mcp/run-docker-mcp.sh"
}
}
}
Docker Installation (HTTP transport):
{
"prestashop": {
"type": "http",
"url": "http://localhost:8765/mcp"
}
}
(Requires running docker compose up -d first)
With VS Code
Create .mcp.json in your project root:
{
"mcpServers": {
"prestashop": {
"type": "stdio",
"command": "prestashop-mcp"
}
}
}
With Claude Code
Create .mcp.json in your project root:
{
"mcpServers": {
"prestashop": {
"command": "/absolute/path/to/run-docker-mcp.sh"
}
}
}
Or for local Python installation:
{
"mcpServers": {
"prestashop": {
"command": "python",
"args": ["-m", "prestashop_mcp.server"]
}
}
}
With Cursor
Add to Cursor's MCP configuration (Settings > Features > MCP Servers):
{
"mcpServers": {
"prestashop": {
"command": "prestashop-mcp"
}
}
}
🛠️ Available MCP Tools
Once configured, your AI assistant has access to 7 specialized tools:
Documentation Tools
1. search_prestashop_docs(query, doc_type, category)
Search ALL PrestaShop documentation using full-text search.
Parameters:
query- Search termsdoc_type- Filter by: hook, guide, tutorial, api, component, etc.category- Filter by: basics, development, modules, themes, etc.
Example:
"Search PrestaShop docs for module development"
"Find installation guides for Mac"
2. get_prestashop_doc(path)
Get the full content of a specific documentation file.
Example:
"Get the Mac installation guide"
3. list_prestashop_docs(doc_type, category)
List available documentation files with optional filters.
Example:
"List all PrestaShop guides"
"Show me all tutorial documents"
4. get_prestashop_stats()
Get statistics about the indexed documentation.
Example:
"Show me PrestaShop documentation statistics"
Hook Tools
5. search_prestashop_hooks(queries, hook_type, origin)
Search PrestaShop hooks specifically.
Example:
"Find all hooks related to products"
6. get_prestashop_hook(hook_name)
Get complete documentation for a specific hook.
Example:
"Show me the displayHeader hook"
7. list_prestashop_hooks(hook_type, origin)
List all hooks organized by type and origin.
Example:
"List all display hooks"
💬 Example Queries
Try asking your AI assistant:
Installation & Setup:
- "How do I install PrestaShop locally on Mac?"
- "What are the system requirements for PrestaShop?"
- "Show me the Docker installation guide"
Module Development:
- "How do I create a PrestaShop module?"
- "Search for module development tutorials"
- "What's the module structure in PrestaShop?"
Theme Development:
- "How do I develop a PrestaShop theme?"
- "Show me Twig template documentation"
- "Find theme development guides"
Hooks:
- "Show me documentation for the displayHeader hook"
- "Find all hooks related to products"
- "List all display hooks"
- "How does the actionProductAdd hook work?"
Components & Architecture:
- "What UI components are available in PrestaShop?"
- "Show me form types documentation"
- "Explain the CQRS pattern in PrestaShop"
APIs:
- "How do I use the PrestaShop Admin API?"
- "Show me webservice documentation"
- "Find API endpoint references"
General:
- "Show me PrestaShop documentation statistics"
- "List all available guides"
- "Search for deployment documentation"
📊 Documentation Coverage
Total: 1,095+ Indexed Documents
By Document Type
- 647 Hooks - All PrestaShop hooks with metadata
- Display hooks (render content)
- Action hooks (trigger functionality)
- Core, Module, and Theme hooks
- Back office and Front office locations
- 138 General Docs - Guides, tutorials, contribution docs
- 119 Components - Forms, Grids, Services, UI components
- 96 API References - Admin API, webservice, domain references
- 64 References - Technical references and specifications
- 12 Guides - Installation, deployment, development guides
- 10 FAQs - Frequently asked questions
- 9 Tutorials - Step-by-step tutorials
By Category
- 697 Module docs - Module development and hooks
- 208 Development docs - Components, architecture, patterns
- 88 Webservice docs - API and webservice references
- 29 Theme docs - Theme development and templates
- 22 Contribution docs - Contributing to PrestaShop
- 17 Basics - Installation, system requirements, deployment
- And more...
Specialized Data
- 386 Domain References - CQRS commands and queries
- 119 UI Components - Form types, grid columns, etc.
🔧 Development
Running Tests
# Run all tests
pytest
# Run specific test
pytest tests/test_ingest.py
# Run with coverage
pytest --cov=prestashop_mcp
Manual Testing
# Index documentation (uses new comprehensive indexer)
python -m prestashop_mcp.ingest_v2
# Force reindex
python -m prestashop_mcp.ingest_v2 --force
# Check database
sqlite3 ~/.mcp/prestashop-docs/database.db "SELECT COUNT(*) FROM prestashop_docs;"
# Get statistics by type
sqlite3 ~/.mcp/prestashop-docs/database.db "SELECT doc_type, COUNT(*) FROM prestashop_docs GROUP BY doc_type;"
📖 How It Works
- Indexing: On first run, indexes 1,095+ documentation files into SQLite with FTS5
- Hooks, guides, tutorials, API docs, components, and more
- Automatic categorization and metadata extraction
- CQRS domain references and UI component detection
- Storage: Documentation stored in
~/.mcp/prestashop-docs/database.db - Search: Uses SQLite FTS5 for efficient full-text search with ranking
- Porter stemming for better search results
- Filter by document type, category, and other metadata
- Retrieval: Direct lookups by path or name for instant access
- Multi-Transport: Supports STDIO (local), HTTP, and SSE (remote) protocols
🤝 Contributing
Contributions welcome! Please read our Contributing Guide.
Quick Start
# Fork and clone
git clone https://github.com/florinel-chis/prestashop-mcp.git
cd prestashop-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
# Format code
black prestashop_mcp/
# Make changes and submit PR
📜 License
MIT License - see LICENSE file for details.
🙏 Acknowledgments
- Inspired by Shopify Liquid MCP
- Docker integration based on best practices from Magento GraphQL Docs MCP, thank you @cmuench
- Built with FastMCP
- Documentation from PrestaShop Docs
- MCP Protocol by Anthropic
🔗 Related Projects
- Shopify Liquid MCP - For Shopify theme development
- Magento GraphQL Docs MCP - For Magento development
- FastMCP - MCP server framework
- Model Context Protocol - MCP specification
📞 Support
- 🐛 Issues
- 💬 Discussions
Made with ❤️ for the PrestaShop development community
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。