Rentalot MCP Server
MCP server for the Rentalot API. Manage rental properties, contacts, showings, conversations, and more from any AI assistant.
README
@rentalot/mcp-server
MCP server for the Rentalot API. Manage rental properties, contacts, showings, conversations, and more from any AI assistant.
Setup
Claude Code
claude mcp add rentalot -e RENTALOT_API_KEY=ra_your_key -- npx -y @rentalot/mcp-server
Codex (OpenAI)
codex mcp add --env RENTALOT_API_KEY=ra_your_key -- npx -y @rentalot/mcp-server
Gemini CLI
gemini mcp add --transport stdio rentalot -- npx -y @rentalot/mcp-server
Then add the env var to ~/.gemini/settings.json:
{
"mcpServers": {
"rentalot": {
"command": "npx",
"args": ["-y", "@rentalot/mcp-server"],
"env": { "RENTALOT_API_KEY": "ra_your_key" }
}
}
}
Claude Desktop / Cursor / Windsurf
All three use the same JSON format — just different file paths:
| Client | Config file |
|---|---|
| Claude Desktop | ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) |
| Cursor | .cursor/mcp.json (project) or ~/.cursor/mcp.json (global) |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
{
"mcpServers": {
"rentalot": {
"command": "npx",
"args": ["-y", "@rentalot/mcp-server"],
"env": {
"RENTALOT_API_KEY": "ra_your_key"
}
}
}
}
OpenCode
Add to ~/.config/opencode/opencode.json:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"rentalot": {
"type": "local",
"command": ["npx", "-y", "@rentalot/mcp-server"],
"environment": { "RENTALOT_API_KEY": "ra_your_key" },
"enabled": true
}
}
}
Get Your API Key
- Go to Rentalot API Keys
- Click Create API Key
- Copy the key (starts with
ra_)
Authentication
You can provide your API key in two ways. Environment variables take priority over the config file.
Option A: Config file (recommended)
Save your key once and every MCP client picks it up automatically — no need to set RENTALOT_API_KEY in each client's config:
mkdir -p ~/.config/rentalot
cat > ~/.config/rentalot/config.yaml << 'EOF'
api_key: ra_your_key
EOF
chmod 600 ~/.config/rentalot/config.yaml
Then register the server without any env var:
# Claude Code
claude mcp add rentalot -- npx -y @rentalot/mcp-server
# Codex
codex mcp add -- npx -y @rentalot/mcp-server
Option B: Environment variable
Pass RENTALOT_API_KEY in your MCP client config (shown in the setup examples above). This overrides the config file.
Config file reference (~/.config/rentalot/config.yaml):
# Required
api_key: ra_your_key
# Optional — defaults to https://rentalot.ai
# base_url: http://localhost:3000
Respects
$XDG_CONFIG_HOME— if set, looks for$XDG_CONFIG_HOME/rentalot/config.yamlinstead.
Resolution order: RENTALOT_API_KEY env var > api_key in config file
AI Agent Skill
This project ships a SKILL.md following the Agent Skills open standard. Install it so your coding agent knows all 65 tools, common workflows, and API patterns.
Quick install with skills CLI (by Vercel Labs):
npx skills add ariel-frischer/rentalot-mcp
<details> <summary><strong>Manual install</strong></summary>
Claude Code — Skills live in ~/.claude/skills/ (global) or .claude/skills/ (project-local).
# Global — available in all projects
mkdir -p ~/.claude/skills/rentalot-mcp
curl -fsSL https://raw.githubusercontent.com/ariel-frischer/rentalot-mcp/main/.skills/default/SKILL.md \
-o ~/.claude/skills/rentalot-mcp/SKILL.md
# Project-local — checked into this repo only
mkdir -p .claude/skills/rentalot-mcp
curl -fsSL https://raw.githubusercontent.com/ariel-frischer/rentalot-mcp/main/.skills/default/SKILL.md \
-o .claude/skills/rentalot-mcp/SKILL.md
Codex CLI — reads skills from ~/.codex/skills/ (global) or .codex/skills/ (project-local).
# Global
mkdir -p ~/.codex/skills/rentalot-mcp
curl -fsSL https://raw.githubusercontent.com/ariel-frischer/rentalot-mcp/main/.skills/default/SKILL.md \
-o ~/.codex/skills/rentalot-mcp/SKILL.md
# Project-local
mkdir -p .codex/skills/rentalot-mcp
curl -fsSL https://raw.githubusercontent.com/ariel-frischer/rentalot-mcp/main/.skills/default/SKILL.md \
-o .codex/skills/rentalot-mcp/SKILL.md
Or pass directly: codex --instructions .skills/default/SKILL.md
</details>
Detailed API Skill
For deeper API reference (full field lists, status enums, webhook events), there's also a comprehensive skill at skills/rentalot/SKILL.md. Symlink it for Claude Code:
ln -s "$(pwd)/skills/rentalot" ~/.claude/skills/rentalot
Then invoke with /rentalot in any Claude Code session.
Tools (65)
| Resource | Tools | Access |
|---|---|---|
| Properties (5) | list_properties, get_property, create_property, update_property, delete_property |
Read: Starter+ / Write: Pro+ |
| Contacts (5) | list_contacts, get_contact, create_contact, update_contact, delete_contact |
Read: Starter+ / Write: Pro+ |
| Showings (6) | list_showings, get_showing, create_showing, update_showing, delete_showing, check_showing_availability |
Read: Starter+ / Write: Pro+ |
| Events (1) | list_events |
Read: Starter+ |
| Conversations (4) | list_conversations, get_conversation, search_conversations, list_conversation_messages |
Read: Starter+ |
| Messages (1) | send_message |
Write: Pro+ |
| Drafts (6) | list_drafts, get_draft, create_draft, update_draft, send_draft, delete_draft |
Read: Starter+ / Write: Pro+ |
| Follow-ups (4) | list_followups, get_followup, create_followup, delete_followup |
Read: Starter+ / Write: Pro+ |
| Workflows (8) | list_workflows, get_workflow, create_workflow, update_workflow, delete_workflow, trigger_workflow_run, list_workflow_runs, get_workflow_run |
Read: Starter+ / Write: Pro+ |
| Webhooks (7) | list_webhooks, create_webhook, get_webhook, update_webhook, delete_webhook, test_webhook, rotate_webhook_secret |
Pro+ |
| Settings (4) | get_settings, update_settings, get_followup_settings, update_followup_settings |
Read: Starter+ / Write: Pro+ |
| Property Images (9) | list_property_images, presign_image_upload, confirm_image_upload, delete_property_images, reorder_property_images, presign_image_batch, confirm_image_batch, import_property_images, get_image_import_job |
Read: Starter+ / Write: Pro+ |
| Sessions (3) | list_sessions, get_session, review_session |
Read: Starter+ / Write: Pro+ |
| Bulk Import (2) | bulk_create_properties, get_bulk_import_job |
Pro+ |
Resources
docs://api-reference— Full API reference (authentication, rate limits, pagination, errors, all resource schemas)
Development
bun install
bun run build # Compile TypeScript
bun run dev # Run with --watch (hot reload)
bun run lint # ESLint
bun run typecheck # tsc --noEmit
bun run test:e2e # E2E test all 65 tools against local dev server
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。