expertapp-receptionist-mcp

expertapp-receptionist-mcp

An MCP server that enables AI agents to act as a business receptionist on expert.com.jm, managing appointments and services without exposing API keys to the model.

Category
访问服务器

README

expertapp-receptionist-mcp

An MCP server that lets an AI agent act as the receptionist for a business on expert.com.jmwithout the API key ever entering the model's own context.

This is the recommended way to connect an agent to expert.com.jm's Reception API. The companion expertapp-receptionist skill (a plain REST API + prompt instructions) works too, but it requires the model itself to construct an Authorization: Bearer <key> header on every request — meaning the raw key has to pass through the model's own reasoning/output at some point, which is exposed to conversation transcripts, logs, and prompt-injection-triggered echoes. This server closes that gap: the key lives only in this process's environment, set once in your agent's own configuration file, never typed into a conversation. The model just calls typed tools (create_booking, check_availability, ...) and never sees a header, a key, or anything that looks like one.

Install & configure

Add it to your MCP client's config (Claude Code, Claude Desktop, Cursor, Windsurf, and most other MCP-compatible tools use a similar JSON config format — check your tool's docs for the exact file/location):

{
  "mcpServers": {
    "expertapp-receptionist": {
      "command": "npx",
      "args": ["-y", "github:incrhst/expertapp-receptionist-mcp"],
      "env": {
        "EXPERTAPP_API_KEY": "your-key-here"
      }
    }
  }
}

Get the key from the business's expert.com.jm dashboard: Team → Reception & management access → Generate an API key for an AI agent. It's shown once — paste it directly into this config file, not into a chat with the agent.

Optional: set EXPERTAPP_BASE_URL in the same env block if the business uses a white-labelled or staging domain. Defaults to https://www.expert.com.jmuse www, not the bare apex domain: the apex 308-redirects to www, and most HTTP clients (correctly, per the fetch spec) strip the Authorization header on that cross-origin hop, which makes requests against the apex silently unauthenticated rather than failing loudly.

Tools

Tool What it does
list_services Every bookable service across the team, labelled with who it belongs to
check_availability Open start times for one service over the next two weeks
create_booking Books an appointment. Paid bookings confirm on payment; free ones hold the slot and require the client to confirm by email
list_bookings Every booking across the team, most recent first
reschedule_booking Moves a booking — requires the client's email/phone to match the record
request_cancellation Queues a cancellation — the booking stays active until the client confirms by email, and may be rejected if it's within the business's cancellation-notice window

Full behavior (including the email-confirmation flows and cancellation policy) is documented in the skill's SKILL.md — this server calls the exact same API, just without the model ever touching the credential.

Scope, on purpose

Same as the REST skill: view services/availability, create/reschedule/cancel bookings. No pricing, subscription/billing, or payout access, and no refund tool — those stay human-only, in the business's own dashboard.

Local development

npm install
npm run build
EXPERTAPP_API_KEY=... node dist/index.js

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选