MarkdownPointer
Enables AI to open, navigate, and export Markdown documents, and to import/export PPTX files.
README
MarkdownPointer
Vibe writing for Markdown. Point at anything, tell AI to fix it.
MarkdownPointer renders your Markdown and lets you click any element - headings, code blocks, table cells, Mermaid diagram nodes, KaTeX math - to copy a filepath:line reference. Paste it into your AI prompt, and the AI knows exactly where to look.
<div align="center"> <img width="640" alt="social-image" src="https://github.com/user-attachments/assets/cdae3548-1e23-4639-9b38-3e03c5c2a337" /> </div>
To change a node's color:
<p align="center"> <img width="40%" valign="middle" alt="Image" src="https://github.com/user-attachments/assets/af06577f-4f7c-40cd-b76e-e3426fd36699" /> <img width="8%" valign="middle" alt="Image" src="https://github.com/user-attachments/assets/16b7d1ab-3799-421a-8d21-9ecdbc1ae66c" /> <img width="40%" valign="middle" alt="Image" src="https://github.com/user-attachments/assets/daea5945-af62-47f7-8f02-a0206611d8e7" /> </p>
Click the node, paste the reference into your prompt, and ask the AI — done.
Color this node orange [c:\docs\architecture.md:6] mermaid node: mdp.exe
More prompt examples:
- Verify this section for technical accuracy [ref]
- Swap these two sections [ref] [ref]
- Delete this [ref]
- Simplify this paragraph [ref]
- Add a code example after this section [ref]
- Fix the grammar here [ref]
- Translate this section to Japanese [ref]
Features
| Feature | Description |
|---|---|
| Point & Prompt | Click any rendered element to copy filepath:line to clipboard |
| Mermaid Diagrams | Flowchart, Sequence, Class, State, ER, Gantt, Pie, Git graph, Mindmap |
| KaTeX Math | Inline $...$ and block $$...$$ |
| SVG | Embedded font support |
| Recent Files | Quick access with pin support |
| Tab Dock/Undock | Drag tabs between windows or detach to a new window |
| Always on Top | Pin the window above other apps for reference |
| Live Reload | Auto-refresh on file changes |
| Export | .pptx (built-in Open XML), .docx (via Pandoc). Mermaid/SVG rendered as images |
| MCP Server | Let AI open, navigate, export documents, and generate/import PPTX |
Install
In a PowerShell 7 console:
Install-Module MarkdownPointer
Quick Start
mdp .\README.md # Open a file
mdp .\docs\*.md # Open multiple files
mdp # Just launch the viewer
MCP Server Setup
Connect MarkdownPointer to Claude so your AI can open and navigate documents directly.
Run these in PowerShell 7:
Register-MdpToClaudeCode # Claude Code
Register-MdpToClaudeDesktop # Claude Desktop
Other MCP clients: Add the output of
Get-MarkdownPointerMCPPath -Escapeas the command in your MCP client's configuration file.
Troubleshooting: If the MCP client fails to connect to the server, .NET 10 Desktop Runtime may not be installed. .NET 10 is not included with Windows — install it and try again.
Then just ask Claude:
- "open README.md in mdp"
- "show the report in mdp and scroll to line 50"
- "export report.md to docx"
- "export slides.md to pptx"
- "import presentation.pptx to markdown"
- "show me slide 3 of slides.md"
MCP Tools
| Tool | Description |
|---|---|
show_markdown |
Open files and scroll to a line |
get_status |
Get current window/tab state |
slide_control |
Navigate reveal.js slides |
get_slide_info |
Get slide shapes and content as text |
get_slide_image |
Get a slide as PNG image (requires PowerPoint) |
export_document |
Export to .pptx (built-in) or .docx (Pandoc) |
import_document |
Import .docx/.pptx to Markdown + extract images |
tag_asset |
Tag imported files and images in index.json |
Keyboard Shortcuts
| Shortcut | Action |
|---|---|
Ctrl+O |
Open file |
Ctrl+F |
Find in page |
Ctrl+G |
Go to line |
Ctrl+P |
|
Ctrl+W / Ctrl+F4 |
Close tab |
Ctrl+Tab / Ctrl+Shift+Tab |
Switch tabs |
Mouse Wheel |
Scroll |
Ctrl+Mouse Wheel |
Zoom |
F5 |
Reload |
Requirements
- Windows 10/11
- PowerShell 7.4+
- .NET 10 Desktop Runtime
<details> <summary>Build from Source</summary>
git clone https://github.com/yotsuda/MarkdownPointer.git
cd MarkdownPointer
.\Build-Deploy.ps1
</details>
License
MIT
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。