Searchspring Integration Assistant

A Model Context Protocol (MCP) server that provides implementation guidance, code validation, and troubleshooting for Searchspring's e-commerce APIs.

Important: This is an integration assistant, not an API proxy. It returns implementation guidance and code examples rather than live API data.

Quick Start for Claude Desktop

1. Install and build:

   git clone <repo-url>
   cd searchspring-api-mcp
   npm install && npm run build

2. Configure Claude Desktop - Add to your claude_desktop_config.json:

   {
     "mcpServers": {
       "searchspring": {
         "command": "node",
         "args": ["/absolute/path/to/searchspring-api-mcp/dist/index.js"]
       }
     }
   }

3. Restart Claude Desktop and ask:

   "Show me how to implement Searchspring search API"

The Searchspring tools will activate automatically - no special commands needed!

Related MCP server: Kintone Development Support MCP Server

Developer Setup

1. Install dependencies:

   npm install
   npm run build

2. Start the server:

   npm start

Note: SEARCHSPRING_SITE_ID is optional - the LLM can provide example site IDs or ask users for their specific ID when needed.

What This MCP Does

✅ Implementation Guidance - Step-by-step API integration instructions for all 8 Searchspring APIs

✅ Code Validation - Analyze existing implementations for issues with platform-specific checks

✅ Platform-Specific Code - Generate code for Shopify, BigCommerce, Magento, and 7+ platforms

✅ Troubleshooting - Diagnose common integration problems with targeted solutions

✅ Modern Platform Support - Includes Shopify checkout extensibility and Web Pixel guidance

✅ Documentation Links - Direct links to relevant Searchspring docs

❌ Not an API Proxy - Does not make live API calls or return product data

Available Tools

🎯 Implementation Guidance

Supported APIs: search, autocomplete, suggest, trending, recommendations, finder, beacon, bulk-index

🔧 Code Generation & Validation

Supported Platforms: shopify, bigcommerce, magento1, magento2, miva, commercev3, 3dcart, volusion, javascript, php, python, custom Supported Code Types: search, autocomplete, suggest, trending, recommendations, finder, beacon, bulk-index, tracking

Example Usage

Get API Implementation Guide

// Get comprehensive guidance for any API
Input: {"api": "search"}
Output: Complete implementation guide with endpoints, examples, and best practices

Get Parameter Details

// Understand specific API parameters
Input: {"api": "search", "parameter": "filters"}
Output: Detailed explanation of filters parameter with examples and best practices

Generate Platform Code

// Generate platform-specific implementation
Input: {"api": "beacon", "platform": "shopify", "eventType": "product"}
Output: Ready-to-use Shopify tracking code with Liquid template syntax

Validate Implementation

// Validate existing code and get troubleshooting help
Input: {
  "code": "<script>ss.track.product.view({sku: 'ABC'});</script>",
  "codeType": "tracking",
  "platform": "shopify",
  "issue": "Events not showing in analytics"
}
Output: ❌ Missing IntelliSuggest script, ⚠️ No safety check, 🔧 Troubleshooting steps

Integration Workflow

1. Planning → Use searchspring_api_guide to understand API structure and requirements

2. Deep Dive → Use searchspring_parameter_guide for specific parameter details

3. Implementation → Use searchspring_code_generator to create platform-specific code

4. Validation → Use searchspring_code_validator to check your implementation

5. Troubleshooting → Use validator with specific issues for detailed diagnosis

Testing & Development

Basic Validation

npm test

Local Development with LLM Clients

Claude Desktop Integration

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "searchspring": {
      "command": "node",
      "args": ["/absolute/path/to/searchspring-api-mcp/dist/index.js"]
    }
  }
}

Testing with OpenAI/Other LLMs

Use any MCP-compatible client:

# Start the MCP server
npm start

# Test with mcp-client or similar tools
npx @modelcontextprotocol/inspector

Manual Tool Testing

Test individual tools directly:

# Example: Test API guide tool
echo '{"method": "tools/call", "params": {"name": "searchspring_api_guide", "arguments": {"api": "search"}}}' | npm start

MCP Activation & Usage

Once configured in Claude Desktop, the Searchspring MCP tools are automatically available. No special activation command needed!

Available Tools:

• searchspring_api_guide - Get implementation guidance for any API

• searchspring_parameter_guide - Get detailed parameter explanations

• searchspring_code_generator - Generate platform-specific code

• searchspring_code_validator - Validate and troubleshoot existing code

Sample Prompts to Activate Tools:

"Show me how to implement Searchspring search API"
"Generate Shopify tracking code for product views"
"Explain the filters parameter for the search API"
"Validate this search implementation code: [paste code]"
"Create BigCommerce autocomplete code"
"How do I implement bulk indexing for Magento?"

Interactive Development Testing

1. Start development server:

   npm run dev

2. Test API guidance:

   • Ask Claude: "Show me how to implement Searchspring search API"

   • Ask Claude: "Generate Shopify tracking code for product views"

   • Ask Claude: "Validate this search implementation code: [paste code]"

3. Test platform-specific generation:

   • Test all platforms: Shopify, BigCommerce, Magento1/2, Miva, CommerceV3, 3dCart, Volusion

   • Test modern Shopify Web Pixel tracking scenarios

   • Test bulk indexing guidance (with and without secret key)

4. Validate cross-references:

   • Ensure all generated code references match docs.searchspring.com

   • Verify platform-specific template syntax (Liquid, Handlebars, PHP)

   • Check that troubleshooting advice matches Zendesk Knowledge Base

Configuration Options

Docker Deployment

Ready for production deployment with Docker:

# Build the Docker image
docker build -t searchspring-mcp .

# Run without environment variables (LLM will handle site IDs)
docker run searchspring-mcp

# Or with optional environment variables
docker run -e SEARCHSPRING_SITE_ID=your_site_id searchspring-mcp

Security Features: Non-root user, minimal Alpine base, production-optimized

Kubernetes Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: searchspring-mcp
spec:
  replicas: 2
  selector:
    matchLabels:
      app: searchspring-mcp
  template:
    metadata:
      labels:
        app: searchspring-mcp
    spec:
      containers:
      - name: searchspring-mcp
        image: searchspring-mcp:latest
        ports:
        - containerPort: 3000
        # Environment variables are optional
        env:
        - name: SEARCHSPRING_TIMEOUT
          value: "10000"

Common Use Cases

New Searchspring Customer: Get implementation guidance for search, autocomplete, and tracking Existing Implementation Issues: Validate code and get troubleshooting help Platform Migration: Generate platform-specific tracking code Development Team Onboarding: Understand API structure and best practices Modern Shopify Stores: Get guidance for checkout extensibility and Web Pixel tracking

Support

• 📖 Documentation: https://docs.searchspring.com/

• 🎯 Help Center: https://help.searchspring.net/

License

MIT