MCP PropTech
Enables AI agents to automate property maintenance workflows by managing tickets, sending WhatsApp notifications via Twilio, and triggering n8n automation webhooks. It allows users to query maintenance statuses, notify tenants, and escalate issues to vendors using natural language.
README
MCP PropTech — Property Maintenance Automation
A production-ready Model Context Protocol (MCP) server that lets a Claude-powered agent orchestrate your full property maintenance workflow — built on top of an existing n8n + WhatsApp automation system.
┌─────────────────────────────────────────────────────────┐
│ Claude Agent / Desktop │
└────────────────────────┬────────────────────────────────┘
│ MCP (stdio / SSE)
┌────────────────────────▼────────────────────────────────┐
│ MCP PropTech Maintenance Server │
│ │
│ get_open_tickets update_maintenance_status │
│ notify_tenant escalate_to_vendor │
└──────┬──────────────────────────────────┬───────────────┘
│ │
┌──────▼──────┐ ┌────────▼──────────┐
│ Ticket DB │ │ Integrations │
│ (in-memory │ │ ┌─────────────┐ │
│ → Postgres │ │ │ n8n │ │
│ in prod) │ │ │ Webhooks │ │
└─────────────┘ │ ├─────────────┤ │
│ │ Twilio │ │
│ │ WhatsApp │ │
│ └─────────────┘ │
└───────────────────┘
Tools
| Tool | Description |
|---|---|
get_open_tickets |
Query tickets filtered by status, priority, category, property |
update_maintenance_status |
Drive tickets through the lifecycle with audit notes |
notify_tenant |
Send WhatsApp messages via 5 templates (acknowledgement, scheduled, update, resolved, delay) |
escalate_to_vendor |
Assign vendor, trigger n8n workflow, notify tenant in one call |
Quick start
# 1. Install dependencies
npm install
# 2. Configure environment
cp .env.example .env
# Fill in ANTHROPIC_API_KEY, Twilio creds, n8n webhook URLs
# 3. Run the agent (single pass over the ticket queue)
npm run agent
# 4. Run as a daemon (checks every 5 minutes)
npm run agent:dev -- --daemon
Claude Desktop integration
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"proptech-maintenance": {
"command": "npx",
"args": ["tsx", "/absolute/path/to/MCPPropTech/src/mcp-server/index.ts"],
"env": {
"ANTHROPIC_API_KEY": "sk-ant-...",
"TWILIO_ACCOUNT_SID": "AC...",
"TWILIO_AUTH_TOKEN": "...",
"TWILIO_WHATSAPP_FROM": "whatsapp:+14155238886",
"N8N_WEBHOOK_MAINTENANCE": "https://your-n8n.com/webhook/maintenance",
"N8N_WEBHOOK_ESCALATE": "https://your-n8n.com/webhook/escalate"
}
}
}
}
Then from Claude Desktop you can say things like:
- "Show me all open emergency tickets"
- "Escalate the burst pipe in unit 3B to a plumber"
- "Send Marcus in 7A a WhatsApp update on his AC ticket"
Architecture
MCP Server (src/mcp-server/)
The server runs over stdio (for Claude Desktop / agent SDK) and exposes the
four tools above. Swap StdioServerTransport for SSEServerTransport to serve
over HTTP for remote deployments.
Ticket Store (src/data/)
An in-memory store with 6 seed tickets covering every status and priority.
Replace TicketStore internals with a Postgres/Supabase/Airtable adapter for
production — the tool handlers only call the store interface.
n8n Integration (src/integrations/n8n.ts)
Fires webhooks to two n8n endpoints:
- Maintenance webhook — on every status change and tenant notification
- Escalation webhook — when a vendor is assigned
Your n8n workflows handle the rest: emailing vendors, updating spreadsheets, creating calendar invites, sending owner reports.
WhatsApp / Twilio (src/integrations/whatsapp.ts)
Uses the Twilio API to send WhatsApp messages. Gracefully falls back to simulation mode (console logging) when credentials are not set — perfect for demos and local development.
Claude Agent (src/agent/maintenance-agent.ts)
A full agentic loop using @anthropic-ai/sdk that:
- Connects to the MCP server via
StdioClientTransport - Fetches available tools and converts them to the Anthropic tool format
- Runs an agentic loop until Claude stops calling tools
- Logs each tool call and result to the console
Lifecycle state machine
open ──→ in_progress ──→ resolved ──→ closed
│ │ ↑
│ └──→ awaiting_tenant
│ │
└──→ escalated ←────┘
│
└──→ in_progress / resolved
Environment variables
| Variable | Required | Description |
|---|---|---|
ANTHROPIC_API_KEY |
Agent only | Claude API key |
TWILIO_ACCOUNT_SID |
Optional | Twilio account SID for WhatsApp |
TWILIO_AUTH_TOKEN |
Optional | Twilio auth token |
TWILIO_WHATSAPP_FROM |
Optional | Sender WhatsApp number |
N8N_WEBHOOK_MAINTENANCE |
Optional | n8n status-change webhook URL |
N8N_WEBHOOK_ESCALATE |
Optional | n8n vendor escalation webhook URL |
N8N_API_KEY |
Optional | n8n API key (if required by your instance) |
VENDOR_CONTACTS |
Optional | JSON map overriding default vendor directory |
AGENT_INTERVAL_MS |
Optional | Daemon poll interval (default: 300000 = 5 min) |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.