Deadline Tracker MCP Server
MCP 项目 — 截止日期 / 课程表追踪器
MCP 课程大作业(FIT-HCMUS,Kha Do 老师)— 基于幻灯片
MCP — Plug Your Agent Into Everything。
结构
mcp-homework/
├── server/
│ ├── core.js ← logic MCP thật: 3 tools + 1 resource + 1 prompt
│ ├── server.js ← chạy core qua stdio
│ └── http.js ← chạy core qua Streamable HTTP + API key guard + /health
├── agent/
│ ├── host.js ← agent = MCP HOST (load config, merge tools, dispatch, use_skill)
│ └── config.json ← khai báo model + các MCP server cần nối + nơi chứa skills
├── skills/
│ └── deadline-review/SKILL.md ← 1 skill orchestrate tool của server
└── README.mdRelated MCP server: ScienceTokyoLMS-mcp
服务器主题:截止日期 / 课程表追踪器
工具
add_deadline({ title, subject, due })— 添加截止日期list_deadlines({ subject?, onlyPending? })— 列出,支持筛选complete_deadline({ id })— 标记完成,如果 id 错误则返回isError: true
资源:
deadline://list— 完整列表,只读提示词:
plan_my_week— 要求模型根据现有截止日期制定每周计划
分部分运行
1. stdio 服务器(用 Inspector 测试)
npm install
npx @modelcontextprotocol/inspector node server/server.js2. HTTP 服务器(本地,然后部署到公网)
MCP_KEY=$(openssl rand -hex 32) node server/http.js
# test:
npx @modelcontextprotocol/inspector # chọn Streamable HTTP, URL http://localhost:3000/mcp,
# header Authorization: Bearer <MCP_KEY>部署(Render/Fly/Railway...):设置环境变量 MCP_KEY,暴露端口 PORT,
/health 用于平台的健康检查。
3. Agent 主机(同时使用两个服务器 + 技能)
cd agent && npm install
# cần Ollama chạy sẵn model qwen3.5:4b (hoặc đổi model/baseURL trong config.json)
node host.js "deadline tuần này của tôi thế nào?"host.js 将:连接 stdio 服务器 + HTTP 服务器 → 合并工具列表 → 从
skills/ 构建技能索引 → 运行 think/decide/act/observe 循环,当问题
匹配技能描述时自动调用 use_skill,然后调用相应的 MCP 工具。
已自行测试
stdio 服务器冒烟测试:调用全部 3 个工具,读取资源,获取提示词,检查 complete 不存在的 id 时返回
isError— 全部符合预期。HTTP 服务器:缺少 key 的请求 → 401;key 正确 → 握手
initialize成功;/health返回ok。
提交前还需完成
将 HTTP 版本部署到公共平台(Render/Fly/Railway/Cloudflare Workers…), 设置
MCP_KEY,记录 URL + key 供评分人员使用安装 Ollama 并拉取
qwen3.5:4b,用真实模型运行agent/host.js(上面的测试只验证了 MCP 层,未调用 LLM)录制简短演示视频(见下方"提交"部分)
This server cannot be deployed
Related MCP Connectors
MCP server for generating rough-draft project plans from natural-language prompts.
An MCP server for deep research or task groups
- mcpOAuthnet.todoist
Official Todoist MCP server for AI assistants to manage tasks, projects, and workflows.
Related MCP Servers
- FlicenseNot gradedqualityCmaintenanceA task manager MCP server that demonstrates all three MCP primitives (tools, resources, prompts). Enables users to manage tasks, read task summaries and details, and run structured planning/review prompts through natural language.-
- FlicenseAqualityDmaintenanceMCP server for Science Tokyo LMS, enabling course material download, assignment deadline checks, and more via tools like list_courses and get_upcoming_deadlines.71-
- FlicenseCqualityDmaintenanceA simple MCP server demonstrating resources, tools, and prompts using a local task list. It enables reading, creating, and analyzing tasks through MCP.1-
- FlicenseNot gradedqualityDmaintenanceA personal MCP server for managing a folder of markdown notes. It provides tools to search and list notes, resources to access individual notes, and a prompt for weekly summaries.-