Allows AI agents to read and edit Markdown documents, including appending sections.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@json-md-studio Update the version in package.json to 2.0.0"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A local desktop editor for JSON & Markdown
Built-in MCP server — let AI agents read & edit your files directly through a gated, permission-scoped interface. 100% local.
Windows · Tauri v2 · React 19 · TypeScript · Vite ·
Json Markdown Studio is a desktop editor for local files: JSON supports tree view / table view / live syntax validation, Markdown provides WYSIWYG editing, and it includes a built-in MCP server so external AI agents (Codex / Claude Code / Cursor) can operate on documents directly.
A lightweight desktop editor for local JSON & Markdown files — tree/table views, live JSON validation, WYSIWYG Markdown editing, plus a built-in MCP server so external AI agents can drive the editor directly.
JSON — Tree / table / code views, live syntax validation with precise error location, path statistics
Markdown — True WYSIWYG editing — click any rendered block and type, formatting applies live; floating format toolbar
AI Integration (MCP Server) — Gated whitelist interface: external AI agents can read/write files, search, patch JSON by path, append markdown sections — read-only / confirm-before-write / full-access modes
Lightweight Desktop Experience — Frameless custom title bar, fully local, no cloud dependency
Related MCP server: LocalNest MCP
The latest installers are available on the Releases page — choose as needed:
Package | Description |
| NSIS installer (recommended for individual distribution, wizard-based installation) |
| MSI installer (enterprise batch deployment / group policy silent install) |
| Portable version, runs without installation |
Download the latest installer from the Releases page — NSIS setup for regular use, MSI for enterprise deployment, or a portable exe that runs without installation.
Prerequisites: Node.js ≥ 20, Rust (MSVC toolchain, see rustup)
npm install # 安装依赖 install dependencies
npm run tauri dev # 桌面开发模式 desktop dev mode (runs Vite + Rust backend)
npm run tauri build # 构建安装包 build installers (NSIS + MSI)Frontend standalone development:
npm run dev(Vite defaults to port 3000).
In Settings → External AI Agent Integration (MCP Server), select the access mode (read-only / confirm-before-write / full access) and the server starts:
Listen address: http://127.0.0.1:<random port>/mcp
Port discovery file: %APPDATA%\com.jsonmarkdownstudio.app\mcp.port
# 示例：Claude Code / Codex 等 MCP 客户端配置
{
"mcpServers": {
"json-md-studio": {
"url": "http://127.0.0.1:<port>/mcp"
}
}
}Built-in tool examples: app.getState (app state), file.read/setContent/replace/insert/delete (file read/write/edit), search.run (search), json.validate (JSON validation), document.updateValueByPath (update value by JSON path), view.setMode (switch view).
Layer | Technology |
Desktop framework | |
Frontend | React 19 · Vite · TypeScript · TailwindCSS 4 · react-markdown |
Backend | Rust · tiny_http (embedded MCP Server) |
Icons | lucide-react |
Issues and PRs are welcome. Core code lives in src/ (frontend) and src-tauri/ (Rust backend, with src-tauri/src/mcp/ for the MCP server).
MIT
This server cannot be deployed
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
One memory, every AI. A shared, user-owned markdown memory your AI clients read and write over MCP.
An MCP server that provides read access to your cloud storage providers, bank accounts and more.
