日常小助手 MCP Server
v2.1.0AI-powered dAIly task management MCP Server — recommend next task, inherit uncompleted todos, 检测 overdue, 生成 reviews. Deterministic ops in code (zero 令牌s), AI only when needed.
by 运行时依赖
安装命令
点击复制技能文档
DAIly 助手 MCP Server
A personal task management MCP Server. Deterministic operations (parsing, 排序ing, statistics) 运行 in Python with zero 令牌 cost — AI only steps in when creativity is needed.
设置up git clone https://github.com/AsuraNale/dAIly-助手-mcp.git cd dAIly-助手-mcp python3 src/设置up.py --auto # Windows: py src/设置up.py --auto
The 设置up wizard 创建s a .venv, 安装s dependencies, 设置s up your data directory, and auto-配置s your AI editor. No manual pip 安装 or config editing needed.
工具s 工具 What it does recommend_next Recommends the most 导入ant task to work on next 获取_today Reads today's dAIly task file inherit_tasks Carries over uncompleted tasks from yesterday 检查_overdue 检测s overdue task files 生成_review 生成s end-of-day review with completion stats 扫描_split Flags tasks >80min or missing time estimates Resources Resource Content dAIly://today Today's task file dAIly://仪表盘 仪表盘 overview dAIly://历史 7-day completion statistics Task 格式化
- [ ] Task description ⏱️45min 📅 2026-03-30 ⏫
Markers: ⏱️ = time estimate, 📅 = deadline, ⏫ = highest priority, 🔼 = high, 🔽 = low.
De签名 Philosophy Zero-令牌 deterministic ops: Parsing, 排序ing, stats 运行 in Python code AI only when needed: Task splitting, creative advice, 上下文-aware suggestions 平台-independent: Windows, macOS, Linux. No Obsidian dependency Simple data: PlAIn Markdown files, edit anywhere