dewy-resort
Sample application for Enterprise MCP server architecture. Built with Workato Enterprise MCP. Hospitality use case, with customer- and employee-facing agentic experiences.
claude mcp add --transport stdio workato-devs-dewy-resort node server.js \ --env NODE_ENV="production" \ --env WORKATO_API_KEY="your-api-key" \ --env WORKATO_BASE_URL="https://your-workato-base-url"
How to use
Dewy Resort is an enterprise-grade MCP server Demo that showcases a two-tier tool architecture built around orchestrators (high-level workflows) and atomic skills (building blocks). The orchestrators coordinate multi-step workflows to ensure data integrity, security, and fast responses, while atomic skills perform single-responsibility tasks that can be composed by AI agents at runtime. This design yields sub-second to multi-second responses for common scenarios such as guest check-ins, room bookings, and service requests, with robust validation and error handling. To interact with the MCP server, clients trigger orchestrator endpoints (for example, check_in_guest or service_request) which then orchestrate underlying atomic skills across integrated backend systems like Salesforce and Stripe via Workato. The result is validated state transitions, parallelizable reads, ordered updates, and a clean separation of concerns that optimizes both performance and flexibility for edge cases.
How to install
Prerequisites:
- Node.js 20+ and npm/yarn installed
- Python 3.11+ (if any Python-based tooling is used in your setup)
- Access to a Workato-based integration hub or equivalent backend orchestration layer
- Git installed
-
Clone the repository git clone https://github.com/workato-devs/dewy-resort.git cd dewy-resort
-
Install dependencies npm install
or if using pnpm/yarn: pnpm install / yarn install
-
Prepare environment variables Create a local environment file or export variables, for example: export NODE_ENV=production export WORKATO_BASE_URL=https://your-workato-base-url export WORKATO_API_KEY=your-api-key
-
Start the MCP server node server.js
-
Verify startup
- Check logs for a successful startup message indicating the MCP server is listening on the configured port
- Use the exposed orchestrator endpoints (e.g., POST /check_in_guest) to validate behavior
-
Optional: Run in Docker (if a Dockerfile is provided in the repo) docker build -t dewy-resort-mcp . docker run -e NODE_ENV=production -e WORKATO_BASE_URL=https://your-workato-base-url -p 3000:3000 dewy-resort-mcp
Note: The bootstrap script referenced in the README automates environment setup for Node.js and Python tooling in typical environments. If you use the bootstrap script, you can replace the manual steps with the one-liner provided in the Quick Start section of the README.
Additional notes
Tips and considerations:
- The architecture emphasizes zero direct system integrations; all data movement should flow through the Workato integration hub for caching, retries, and centralized security.
- Orchestrators should validate prerequisites in parallel when possible to minimize latency.
- Atomic skills should be designed as single-responsibility building blocks that can be recomposed by agents at runtime.
- Ensure idempotency in orchestrators to avoid duplicate state transitions and simplify retry logic.
- When debugging, inspect both orchestrator sequencing and the individual atomic skill calls to identify where latency or failures occur.
- Populate environment variables securely (do not commit secrets). Use a secrets manager or CI/CD vault in production.
- If you transition to Docker or another deployment target, mirror the environment variables and ports used in development to avoid runtime mismatches.
- Monitor performance: typical orchestrators aim for < 3 seconds response times with 4-6 API calls; adjust timeouts and parallelism as needed for scale.
Related MCP Servers
obsidian -tools
Add Obsidian integrations like semantic search and custom Templater prompts to Claude or any MCP client.
ironcurtain
A secure* runtime for autonomous AI agents. Policy from plain-English constitutions. (*https://ironcurtain.dev)
nutrient-document-engine
A Model Context Protocol (MCP) server implementation exposes document processing capabilities through natural language, supporting both direct human interaction and AI agent tool calling.
create -app
A CLI tool for quickly scaffolding Model Context Protocol (MCP) server applications with TypeScript support and modern development tooling
grok-faf
First MCP server for Grok | FAST⚡️AF • URL-based AI context • Vercel-deployed
agent-rules
MCP server that enables your agents to use coding rules from any or your GitHub repository. Instead of workspace rules files, you can now prompt agents to access the your coding rules from any repository.