TenantCloud MCP Server
Enables AI agents to manage rental properties on TenantCloud via natural language, including contacts, messaging, maintenance, and financials.
README
tenantcloud-client-ts
TypeScript client library and MCP server for TenantCloud, a rental property management platform. A port of yllibed/TenantCloudClient (C#/.NET), extended well beyond parity.
This is not an official TenantCloud product. TenantCloud does not provide a public API; this library works against their internal endpoints and can break whenever TenantCloud changes their frontend. Use at your own risk.
Packages
| Package | npm | Description |
|---|---|---|
packages/client |
tenantcloud-client |
API client: typed sub-clients for messaging, maintenance, financials, and leasing; paginated sources with fluent filters; generic JSON:API CRUD; token stores (OS credential store, file); CDP-based browser auth |
packages/mcp |
tc-mcp |
MCP server (stdio) exposing TenantCloud to AI agents (Claude Desktop, Claude Code, Cursor), with a login/logout/install CLI |
Quick start
Client library
npm install tenantcloud-client
import { TcClient, SecureTokenStore } from "tenantcloud-client";
import { CdpTokenProvider } from "tenantcloud-client/cdp";
const provider = new CdpTokenProvider({
tokenStore: new SecureTokenStore(),
allowInteractiveLogin: true,
});
const tc = new TcClient(provider);
// Reads with fluent filters
const user = await tc.getUserInfo();
const tenants = await tc.contacts.onlyTenants().getAll();
const overdue = await tc.transactions.forCategory("income").forStatus("with_balance").getAll();
// Typed sub-clients
const threads = await tc.messaging.threads("tenant");
await tc.messaging.sendMessage(threads.items[0].id, "Hi!");
const request = await tc.maintenance.create({ title: "Leaky faucet", property_id: 123 });
// First contact with a lead (creates their message thread)
const thread = await tc.leasing.openLeadThread(leadId);
await tc.messaging.sendMessage(thread.id, "Thanks for your interest!");
// Any other endpoint via generic JSON:API CRUD
const tasks = tc.resource("/tasks", "task");
await tasks.create({ title: "Call plumber" });
MCP server
One command (needs Node 20+):
npx tc-mcp install claude-code # or: claude-desktop
Restart Claude Code and ask things like "who owes rent?", "message the lead who inquired yesterday", or "create a maintenance request for unit 3B". The first time, the agent notices you are not signed in and offers to open a TenantCloud sign-in window (normal password + 2FA); tokens go to your OS credential store. To sign in ahead of time instead: npx tc-mcp login.
Hosted server (Claude on web, teams)
For teammates on claude.ai, tc-mcp serve runs a multi-user remote server: OAuth 2.1 in front (claude.ai custom connector flow, email + invite code), an encrypted per-user TenantCloud token vault in Postgres, and per-person pairing via tc-mcp login --remote. Each teammate's tool calls run under their own TenantCloud account and permissions.
Each company hosts its own instance (fork or clone this repo - there is no shared multi-company server by design). Step-by-step Railway guide, including CLI commands and troubleshooting: docs/DEPLOY_RAILWAY.md.
MCP tools
35 tools across six areas, plus resources tc://guide, tc://catalog, tc://property/{id}, tc://unit/{id}, tc://contact/{id}:
- Core reads:
get_user_info,list_contacts,list_properties,list_units,list_transactions,list_leases - Messaging:
list_message_channels,list_threads,find_threads,list_messages,send_message,message_lead(first contact with a lead: creates their thread and sends in one step),mark_thread_read - Maintenance:
list_maintenance_requests,get_maintenance_request,create_maintenance_request,update_maintenance_request,resolve_maintenance_request,list_inspections - Financials:
get_transaction_statistics,create_transaction,update_transaction,delete_transaction,list_recurring_transactions,list_payments,list_reconciliation_accounts,owner_balances - Leasing:
get_lease,update_lease,list_lease_notices,list_applications,list_screenings,list_leads,create_lead - Escape hatch:
tc_request(any of the ~1100 cataloged endpoints; seetc://catalog)
Tool responses resolve foreign-key IDs to human-readable names via an entity cache, and messenger payloads are slimmed so they fit agent context windows.
How authentication works
TenantCloud has no OAuth or API keys. The CDP token provider:
- Returns the in-memory token if its JWT is still valid
- Loads from the token store, refreshing via
POST /auth/tokenif expired - Connects to a running Chromium browser (debug port 9222) and extracts
access_token/fingerprintfrom localStorage and thetc_refresh_tokencookie of anapp.tenantcloud.comtab - If allowed, launches a temporary browser window for interactive login and polls until you finish signing in
Development
npm install
npm run build # tsc --build for both packages
npm test # vitest (unit + MCP in-memory transport tests)
TC_AUTH_TOKEN=<jwt> npm test # also runs live API integration tests
See SPEC.md for design decisions and the documented internal API surface (~1100 endpoints, discovered from the SPA bundle; full list in docs/api-catalog.txt).
License
MIT. Portions derived from yllibed/TenantCloudClient, copyright Carl de Billy, MIT.
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.
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.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.