tagmesh
Enables querying and managing markdown journal entries via tags, full-text search, and CRUD operations, plus workspace statistics, through a native edge MCP interface.
README
<div align="center">
🌸 TagMesh
Zero-Folder · Titleless · Tag-Woven Markdown Notes System with Claymorphic Aesthetics & Serverless Edge
Ditch folders and titles. Weave thoughts organically through #hashtags at the edge.
<br/>
<br/>
</div>
✨ Key Features
- 🏷️ Zero Titles & Zero Folders: Treat notes as fluid Markdown streams. Categorization relies purely on
#hashtagswith automatic first-line excerpt summaries. - 🎨 Tactile Claymorphism Aesthetics: Rounded clay pills, spatial relief shadows, auditory soundscapes, particle bursts, and 6 curated mood themes.
- 💬 Cross-Device Shared Danmaku Plaza: Thoughts sent from mobile fly across desktop monitors in real time with synchronized reaction hearts and authoritative telemetry.
- 🧭 6 Multi-Dimensional Galleries: Bento Grid, 3D Spatial Carousel, Galaxy Mesh Graph, Polaroid Board, Timeline Stream, and Freeform Canvas.
- 🤖 Native Serverless MCP Interface: Edge-native Model Context Protocol (JSON-RPC 2.0) interface with 8 production tools ready for Claude Desktop and Cursor.
⚡ Essential Shortcuts
| Shortcut | Description | Scope |
|---|---|---|
| <kbd>Cmd</kbd> + <kbd>K</kbd> / <kbd>Ctrl</kbd> + <kbd>K</kbd> | Open Global Command Central (Fulltext search & quick create) | Global |
| <kbd>Cmd</kbd> + <kbd></kbd> / <kbd>Ctrl</kbd> + <kbd></kbd> | Toggle Left TagMesh Slide-over Sidebar | Global |
| <kbd>Cmd</kbd> + <kbd>N</kbd> / <kbd>Ctrl</kbd> + <kbd>N</kbd> | Instant Create Blank Note (100ms response) | Global |
| <kbd>#</kbd> | Trigger Hashtag Autocomplete Menu | Editor |
| <kbd>Cmd</kbd> + <kbd>Shift</kbd> + <kbd>L</kbd> | Toggle Bilingual UI (English / 简体中文) | Global |
🚀 Quick Local Development
# 1. Clone repo & install dependencies
git clone https://github.com/SaulGoodManC99/TagMesh.git
cd TagMesh
npm install
# 2. Start frontend dev server (with LAN access on 0.0.0.0:5173)
npm run dev
# 3. Start Cloudflare Worker backend locally
npm run worker:dev
☁️ Cloudflare Deployment Guide
TagMesh runs on Cloudflare Workers + D1 SQLite Database:
1. Create D1 Database
Go to Cloudflare Dashboard -> D1 SQL Database -> Create database (name it tagmesh-db), copy your Database ID, and paste the contents of schema.sql into its Console to initialize tables.
2. Configure GitHub Actions Auto-Deployment (Recommended)
Add 3 secrets in your GitHub repo Settings -> Secrets and variables -> Actions:
CLOUDFLARE_API_TOKEN: Cloudflare API TokenCLOUDFLARE_ACCOUNT_ID: Cloudflare Account IDCLOUDFLARE_D1_DATABASE_ID: D1 Database ID from Step 1
Every
git pushtomainnow automatically builds and deploys to Cloudflare Edge in ~30 seconds!
3. Local CLI Deployment (Optional)
npx wrangler login
npm run db:init:remote
npm run build
npm run worker:deploy
🤖 MCP (Model Context Protocol) Integration
Add to your Claude Desktop configuration (claude_desktop_config.json):
{
"mcpServers": {
"tagmesh": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-fetch",
"https://your-domain.com/mcp"
]
}
}
}
Available tools: search_by_tag, search_fulltext, read_note, create_or_update_note, append_to_note, delete_note, list_tags, get_workspace_stats.
📄 License
Distributed under the MIT License.
<div align="center">
Crafted with 💖 and Clay Magic by SaulGoodManC99
</div>
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。