fix(registry): reconcile catalog with public Studio entries

.github/workflows/checks.yml

@@ -30,7 +30,9 @@ jobs:
         run: bun run lint
 
       - name: Check registry.json is up to date
-        run: bun run check:registry
+        run: |
+          bun test scripts/lib/registry-payload.test.ts
+          bun run check:registry
 
       # Runs over every MCP, not only the changed ones: deleting a withAuth()
       # wrapper elsewhere in the repo must fail the PR that does it.

ai-gateway/app.json

@@ -0,0 +1,33 @@
+{
+  "scopeName": "deco",
+  "name": "ai-gateway",
+  "friendlyName": "Deco AI Gateway",
+  "connection": {
+    "type": "HTTP",
+    "url": "https://sites-deco-ai-gateway.deco.site/mcp"
+  },
+  "description": "LLM gateway with automatic OpenRouter API key provisioning per organization. Each org gets its own isolated key for cost tracking and spending visibility.",
+  "unlisted": false,
+  "icon": "https://assets.decocache.com/mcp/6e1418f7-c962-406b-aceb-137197902709/ai-gateway.png",
+  "metadata": {
+    "categories": [
+      "AI"
+    ],
+    "official": false,
+    "verified": true,
+    "tags": [
+      "llm",
+      "openrouter",
+      "ai",
+      "gateway",
+      "cost-tracking",
+      "multi-model",
+      "gpt",
+      "claude",
+      "gemini",
+      "provisioning"
+    ],
+    "short_description": "LLM gateway with automatic per-org OpenRouter API key provisioning and spending visibility.",
+    "mesh_description": "LLM gateway with automatic OpenRouter API key provisioning per organization. Each org gets its own isolated key for cost tracking and spending visibility."
+  }
+}

apify/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["apify", "scraping", "automation", "actors", "data-extraction", "web-crawling"],
     "short_description": "Run Apify actors for web scraping and automation tasks.",
-    "mesh_description": "The Apify MCP provides comprehensive integration with the Apify platform, enabling AI agents to run web scraping actors, manage automation tasks, and extract data from websites. This MCP allows listing available actors, getting actor details, running actors synchronously or asynchronously, and retrieving run results with dataset items. Perfect for building intelligent automation workflows, data collection pipelines, and web monitoring solutions. Authentication is done via Apify API token passed in the Authorization header."
+    "mesh_description": "The Apify MCP provides comprehensive integration with the Apify platform, enabling AI agents to run web scraping actors, manage automation tasks, and extract data from websites. This MCP allows listing available actors, getting actor details, running actors synchronously or asynchronously, and retrieving run results with dataset items. Perfect for building intelligent automation workflows, data collection pipelines, and web monitoring solutions. Authentication is done via Apify API token passed in the Authorization header.",
+    "verified": false
   }
 }
 

centralize-image/app.json

@@ -0,0 +1,41 @@
+{
+  "scopeName": "deco",
+  "name": "centralize-image",
+  "friendlyName": "Centralize Product Image",
+  "connection": {
+    "type": "HTTP",
+    "url": "https://centralize-image-production.up.railway.app/api/mcp"
+  },
+  "description": "Remove the background from a product image, auto-center it on a configurable canvas, and fine-tune position, scale, and crop.",
+  "unlisted": false,
+  "requester": {
+    "repository": "https://github.com/decocms/centralize-image"
+  },
+  "metadata": {
+    "categories": [],
+    "official": true,
+    "verified": true,
+    "tags": [
+      "image",
+      "product",
+      "background-removal",
+      "mcp-app"
+    ],
+    "short_description": "Background-removal + auto-centering editor for product images.",
+    "mesh_description": "Remove the background from a product image, auto-center it on a configurable canvas, and fine-tune position, scale, and crop."
+  },
+  "tools": [
+    {
+      "name": "open_centralize_editor",
+      "description": "Open the Centralize Product Image editor UI. Lets the user upload an image and iterate on centering parameters."
+    },
+    {
+      "name": "centralize_product_image",
+      "description": "Remove the background from a product image and auto-center it on a W×H canvas with a configurable minimum margin. Returns the final PNG."
+    },
+    {
+      "name": "MCP_CONFIGURATION",
+      "description": "MCP Configuration"
+    }
+  ]
+}

clickhouse/app.json

@@ -14,6 +14,7 @@
     "official": true,
     "tags": ["clickhouse", "database", "analytics", "sql", "olap", "data-warehouse"],
     "short_description": "Query and manage ClickHouse databases with SQL",
-    "mesh_description": "ClickHouse MCP enables AI agents to interact with ClickHouse, the fast columnar database for real-time analytics. Run SQL queries, explore schemas, and analyze billions of rows in seconds. Perfect for log analysis, time-series data, and business intelligence."
+    "mesh_description": "ClickHouse MCP enables AI agents to interact with ClickHouse, the fast columnar database for real-time analytics. Run SQL queries, explore schemas, and analyze billions of rows in seconds. Perfect for log analysis, time-series data, and business intelligence.",
+    "verified": false
   }
 }

cloudflare-ai-gateway/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["cloudflare", "ai", "gateway", "openai", "anthropic", "rate-limiting", "caching", "monitoring"],
     "short_description": "Manage AI API requests with monitoring, caching, rate limiting, and cost control",
-    "mesh_description": "Cloudflare AI Gateway MCP provides unified management and control for AI API requests across multiple providers. This official MCP enables you to route requests through Cloudflare's global network for improved reliability and speed, cache AI responses to reduce costs and latency, implement rate limiting and request quotas per user or application, monitor AI API usage with detailed analytics and logging, implement fallback logic across multiple AI providers, control costs with spending limits and alerts, add authentication and access control to AI endpoints, and anonymize or redact sensitive data in requests. The platform supports all major AI providers including OpenAI, Anthropic, Google, Azure OpenAI, Hugging Face, and more. Perfect for applications using multiple AI models, managing AI costs at scale, ensuring reliability with automatic failover, monitoring AI usage patterns, and implementing governance policies. Use natural language to configure gateways, analyze usage patterns, optimize costs, set rate limits, and troubleshoot API issues across your AI infrastructure."
+    "mesh_description": "Cloudflare AI Gateway MCP provides unified management and control for AI API requests across multiple providers. This official MCP enables you to route requests through Cloudflare's global network for improved reliability and speed, cache AI responses to reduce costs and latency, implement rate limiting and request quotas per user or application, monitor AI API usage with detailed analytics and logging, implement fallback logic across multiple AI providers, control costs with spending limits and alerts, add authentication and access control to AI endpoints, and anonymize or redact sensitive data in requests. The platform supports all major AI providers including OpenAI, Anthropic, Google, Azure OpenAI, Hugging Face, and more. Perfect for applications using multiple AI models, managing AI costs at scale, ensuring reliability with automatic failover, monitoring AI usage patterns, and implementing governance policies. Use natural language to configure gateways, analyze usage patterns, optimize costs, set rate limits, and troubleshoot API issues across your AI infrastructure.",
+    "verified": false
   }
 }
 

cloudflare-auditlogs/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["cloudflare", "audit", "logs", "security", "compliance", "governance", "activity-tracking"],
     "short_description": "Access account audit logs for tracking user actions, changes, and API activity",
-    "mesh_description": "Cloudflare Audit Logs MCP provides comprehensive access to account-level audit trails for security monitoring and compliance requirements. This official MCP enables you to track all user actions and configuration changes across your Cloudflare account, monitor API access and usage patterns, view authentication events (logins, failures, MFA changes), audit permission and role changes for team members, track zone and domain configuration modifications, monitor security policy updates (firewall, rate limiting, bot management), generate compliance reports for SOC 2, ISO 27001, and other frameworks, configure alerts for suspicious or unauthorized activities, and export audit logs for long-term retention and SIEM integration. The platform provides detailed timestamps, user identification, IP addresses, and action descriptions for every event. Perfect for security teams monitoring for threats, compliance officers maintaining audit trails, administrators tracking configuration changes, and organizations requiring detailed activity logging. Use natural language to search audit logs, investigate security incidents, generate compliance reports, identify policy violations, and analyze user behavior patterns across your Cloudflare organization."
+    "mesh_description": "Cloudflare Audit Logs MCP provides comprehensive access to account-level audit trails for security monitoring and compliance requirements. This official MCP enables you to track all user actions and configuration changes across your Cloudflare account, monitor API access and usage patterns, view authentication events (logins, failures, MFA changes), audit permission and role changes for team members, track zone and domain configuration modifications, monitor security policy updates (firewall, rate limiting, bot management), generate compliance reports for SOC 2, ISO 27001, and other frameworks, configure alerts for suspicious or unauthorized activities, and export audit logs for long-term retention and SIEM integration. The platform provides detailed timestamps, user identification, IP addresses, and action descriptions for every event. Perfect for security teams monitoring for threats, compliance officers maintaining audit trails, administrators tracking configuration changes, and organizations requiring detailed activity logging. Use natural language to search audit logs, investigate security incidents, generate compliance reports, identify policy violations, and analyze user behavior patterns across your Cloudflare organization.",
+    "verified": false
   }
 }
 

cloudflare-autorag/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["cloudflare", "rag", "vector-database", "embeddings", "ai", "semantic-search", "llm"],
     "short_description": "Build RAG applications with automatic vector database and embedding management",
-    "mesh_description": "Cloudflare AutoRAG MCP provides automated Retrieval-Augmented Generation infrastructure for building AI applications with context-aware responses. This official MCP enables you to automatically generate and store embeddings for your documents, manage vector databases with semantic search capabilities, integrate with Large Language Models for contextual responses, configure chunking strategies for optimal retrieval, implement hybrid search combining vector and keyword matching, manage knowledge bases with automatic updates and versioning, optimize retrieval performance with caching and indexing, and monitor RAG pipeline performance and accuracy. The platform handles the complexity of vector embeddings, similarity search, and context injection, allowing you to focus on building AI features. Perfect for building chatbots with document knowledge, semantic search engines, question-answering systems, content recommendation engines, and any application requiring AI with access to external knowledge. Integrates seamlessly with Cloudflare AI and Workers. Use natural language to configure RAG pipelines, manage knowledge bases, optimize retrieval accuracy, and troubleshoot context quality issues."
+    "mesh_description": "Cloudflare AutoRAG MCP provides automated Retrieval-Augmented Generation infrastructure for building AI applications with context-aware responses. This official MCP enables you to automatically generate and store embeddings for your documents, manage vector databases with semantic search capabilities, integrate with Large Language Models for contextual responses, configure chunking strategies for optimal retrieval, implement hybrid search combining vector and keyword matching, manage knowledge bases with automatic updates and versioning, optimize retrieval performance with caching and indexing, and monitor RAG pipeline performance and accuracy. The platform handles the complexity of vector embeddings, similarity search, and context injection, allowing you to focus on building AI features. Perfect for building chatbots with document knowledge, semantic search engines, question-answering systems, content recommendation engines, and any application requiring AI with access to external knowledge. Integrates seamlessly with Cloudflare AI and Workers. Use natural language to configure RAG pipelines, manage knowledge bases, optimize retrieval accuracy, and troubleshoot context quality issues.",
+    "verified": false
   }
 }
 

cloudflare-browser/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["cloudflare", "browser", "puppeteer", "playwright", "automation", "scraping", "testing", "rendering"],
     "short_description": "Control and automate cloud browsers with Puppeteer and Playwright for scraping and testing",
-    "mesh_description": "Cloudflare Browser MCP provides cloud-based browser automation and rendering capabilities using Puppeteer and Playwright at the edge. This official MCP enables you to run browser automation scripts without managing infrastructure, scrape dynamic websites with JavaScript rendering, perform automated testing across different browsers and devices, capture screenshots and generate PDFs of web pages, monitor website availability and visual changes, bypass bot detection and anti-scraping measures, execute browser scripts at scale with automatic load balancing, and access websites from global edge locations for localized testing. The platform handles browser lifecycle management, resource cleanup, and provides reliable execution at scale. Perfect for web scraping projects, automated testing pipelines, monitoring services, PDF generation from HTML, and any workflow requiring programmatic browser control. Supports full Puppeteer and Playwright APIs with additional Cloudflare-specific optimizations. Use natural language to create scraping workflows, run automated tests, capture page snapshots, and build browser automation solutions."
+    "mesh_description": "Cloudflare Browser MCP provides cloud-based browser automation and rendering capabilities using Puppeteer and Playwright at the edge. This official MCP enables you to run browser automation scripts without managing infrastructure, scrape dynamic websites with JavaScript rendering, perform automated testing across different browsers and devices, capture screenshots and generate PDFs of web pages, monitor website availability and visual changes, bypass bot detection and anti-scraping measures, execute browser scripts at scale with automatic load balancing, and access websites from global edge locations for localized testing. The platform handles browser lifecycle management, resource cleanup, and provides reliable execution at scale. Perfect for web scraping projects, automated testing pipelines, monitoring services, PDF generation from HTML, and any workflow requiring programmatic browser control. Supports full Puppeteer and Playwright APIs with additional Cloudflare-specific optimizations. Use natural language to create scraping workflows, run automated tests, capture page snapshots, and build browser automation solutions.",
+    "verified": false
   }
 }
 

cloudflare-builds/app.json

@@ -15,7 +15,8 @@
     "official": true,
     "tags": ["cloudflare", "pages", "builds", "deployment", "ci-cd", "static-sites", "jamstack"],
     "short_description": "Manage Cloudflare Pages builds and deployments with build monitoring and configuration",
-    "mesh_description": "Cloudflare Builds MCP provides comprehensive management of Cloudflare Pages build and deployment processes. This official MCP enables you to trigger new builds and deployments for your static sites, monitor build status and progress in real-time, access build logs and error messages, manage build configurations and environment variables, configure custom build commands and output directories, integrate with Git repositories for automatic deployments, manage preview deployments for pull requests, and control build caching for faster deployment times. The platform supports popular frameworks like React, Vue, Next.js, Gatsby, Hugo, and more. Perfect for automating deployment workflows, managing multiple projects, monitoring build health, and integrating CI/CD pipelines with Cloudflare Pages. Use natural language to check build status, troubleshoot failed builds, manage deployment settings, and optimize build performance across your projects."
+    "mesh_description": "Cloudflare Builds MCP provides comprehensive management of Cloudflare Pages build and deployment processes. This official MCP enables you to trigger new builds and deployments for your static sites, monitor build status and progress in real-time, access build logs and error messages, manage build configurations and environment variables, configure custom build commands and output directories, integrate with Git repositories for automatic deployments, manage preview deployments for pull requests, and control build caching for faster deployment times. The platform supports popular frameworks like React, Vue, Next.js, Gatsby, Hugo, and more. Perfect for automating deployment workflows, managing multiple projects, monitoring build health, and integrating CI/CD pipelines with Cloudflare Pages. Use natural language to check build status, troubleshoot failed builds, manage deployment settings, and optimize build performance across your projects.",
+    "verified": false
   }
 }