Internship Tracker MCP Server
Enables Claude to manage internship applications in Supabase by adding, listing, updating, and deleting entries via MCP tools.
README
Internship tracker MCP server
A small backend that exposes your Supabase applications table as MCP tools,
so Claude can add, list, update, and delete entries directly in conversation.
Files
db.js— the only file that talks to Supabase directlytools.js— the four tool definitions (add / list / update / delete), each with a schema (what input is valid) and a handler (what it does to the database)server.js— the HTTP server that speaks the MCP protocol and wires the tools together; this is what Claude actually connects to.env.example— copy to.envand fill in your real Supabase + API key values
Run it locally first
npm install
cp .env.example .env
# edit .env with your real Supabase URL, service key, and a made-up API key
npm run dev
Visit http://localhost:3000 in a browser — you should see
"Internship tracker MCP server is running." That confirms the server itself
is up, before you worry about MCP or Claude at all.
Deploy it somewhere public
Claude's custom connectors are reached from Anthropic's cloud, not your
laptop, so localhost will not work once you're ready to connect it to
Claude. Options: Render, Fly.io, Railway.
General steps for any of them:
- Push this folder to a GitHub repo
- Connect that repo to your chosen host
- Set the same environment variables from
.envin the host's dashboard (never commit the real.envfile) - Deploy — you'll get a public URL like
https://internship-tracker.onrender.com - Your MCP endpoint is that URL +
/mcp, e.g.https://internship-tracker.onrender.com/mcp
Add it to Claude
In Claude settings → Connectors → Add custom connector, paste the /mcp
URL and configure the header x-api-key with the same value you set as
CONNECTOR_API_KEY.
Next steps
- Test each tool individually once connected: "add a test application", "list my applications", "update it to interview stage", "delete it"
- Once the plain tools work, the visual departures-board UI can be added
as an MCP App resource on top of
list_applications— a follow-up step, not required for the core tracker to work
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。