Google Business Profile Review MCP Server

# MCP Resources This MCP server provides read-only resources that can be accessed by AI assistants and other MCP clients to gather context about your Google Business Profile. ## Available Resources ### 1. **Locations** (`locations://list`) Lists all business locations associated with your Google Business Profile account. **URI**: `locations://list` **Returns**: ```json { "totalCount": 1, "locations": [ { "name": "locations/6087465285515339471", "displayName": "Noodle House", "address": { "addressLines": ["Kulathoor, Near NH Service Road"], "locality": "Thiruvananthapuram", "administrativeArea": "Kerala", "postalCode": "695583", "regionCode": "IN" }, "phone": "+91 1234567890", "website": "https://noodlehouse.in/" } ] } ``` **Use Cases**: - Get location IDs for fetching reviews - Display business information - Multi-location management --- ### 2. **Reviews** (`reviews://{locationId}` and `reviews://all`) Fetches unreplied reviews for specific locations or all locations. **Dynamic URI**: `reviews://{locationId}` (replace `{locationId}` with actual location ID) **Convenience URI**: `reviews://all` (all locations at once) **Example URIs**: - `reviews://locations/6087465285515339471` - Reviews for specific location - `reviews://all` - All unreplied reviews across all locations **Returns** (for specific location): ```json { "locationName": "locations/6087465285515339471", "totalReviews": 50, "unrepliedCount": 6, "reviews": [ { "id": "AbFvOqk7ktaqtIRXifV8ph...", "rating": "FIVE", "text": "Great food!", "author": "John Doe", "isAnonymous": false, "createTime": "2025-11-20T10:30:00Z", "updateTime": "2025-11-20T10:30:00Z", "hasReply": false } ] } ``` **Returns** (for all locations): ```json { "totalLocations": 1, "totalUnrepliedReviews": 6, "locations": [ { "location": { "name": "locations/6087465285515339471", "displayName": "Noodle House", "address": {...} }, "reviews": [...], "totalReviews": 50, "unrepliedCount": 6 } ] } ``` **Use Cases**: - Check pending reviews for a specific location (using dynamic URI) - Get overview of all locations' reviews (using reviews://all) - Fetch fresh review data in real-time - Filter reviews by location --- ### 3. **Business Profile** (`business_profile://profile`) Provides enhanced business profile information with response guidelines and brand voice. **URI**: `business_profile://profile` **Returns**: ```json { "name": "locations/6087465285515339471", "locationName": "Noodle House", "address": {...}, "businessType": "Restaurant", "language": "en", "description": "Noodle House", "categories": ["Asian Restaurant", "Noodle Shop"], "responseGuidelines": { "tone": "professional and friendly", "style": "conversational yet business-appropriate", "thankCustomers": true, "addressConcerns": true, "inviteReturn": true, "includeBusinessName": true, "maxLength": 4096, "language": "en" }, "brandVoice": { "personality": "warm, welcoming, food-passionate", "values": [ "customer satisfaction", "quality service", "continuous improvement", "community engagement" ] } } ``` **Use Cases**: - Generate consistent brand voice in responses - Understand business context for better replies - Apply response guidelines automatically --- ### 4. **Review Templates** (`review_templates://templates`) Pre-defined response templates for different review types and scenarios. **URI**: `review_templates://templates` **Returns**: Comprehensive template library organized by: - **Star ratings** (5-star, 4-star, 3-star, 1-2 star) - **Specific scenarios** (food service, retail, service industry) - **Special situations** (no comment, first-time visitor, regular customer) - **Tonal variations** (professional, friendly, warm) **Template Example**: ```json { "positiveReviews": { "fiveStars": [ "Thank you so much for the wonderful 5-star review, {customerName}! We're thrilled to hear about your positive experience with {businessName}..." ] } } ``` **Placeholders**: - `{customerName}` - Reviewer's display name - `{businessName}` - Your business name - `{starRating}` - Star rating (1-5) - `{reviewText}` - Original review text - `{date}` - Current date - `{location}` - Business location/address **Use Cases**: - Quick response generation - Consistent messaging - Training data for AI responses - Fallback when LLM is unavailable --- ## How to Use Resources ### In Prompts Resources can be referenced in prompts to provide context: ```typescript // Example: Using locations resource in a prompt const locations = await client.readResource('locations://list'); const prompt = `Based on these locations: ${locations}...`; ``` ### In AI Assistants AI assistants with MCP support can automatically access these resources to: - Understand your business context - Generate appropriate responses - Stay consistent with your brand voice ### Direct Access ```bash # Using MCP CLI (if available) mcp read-resource "reviews://all" mcp read-resource "locations://list" ``` --- ## Benefits of Resource-Based Architecture 1. **Read-Only Access**: Resources are safe to expose - they only read data, never modify 2. **Cacheable**: Resource responses can be cached for performance 3. **Context-Rich**: Provides comprehensive context without requiring multiple tool calls 4. **Standardized**: MCP standard format for interoperability 5. **Discoverable**: Clients can list available resources automatically --- ## Next Steps - **Enhance Prompts**: Update review response prompts to reference these resources - **Cache Strategy**: Implement caching for frequently accessed resources - **Dynamic Resources**: Add support for per-location review resources - **Webhooks**: Consider webhook-triggered resource updates for real-time data