Plaud MCP Server
Enables AI assistants to access Plaud Note recordings, providing direct access to speaker-diarized transcripts and AI-generated meeting summaries. Users can search, list, and retrieve recording details directly within MCP-compatible clients like Claude and Cursor.
README
Plaud MCP Server
Connect your Plaud Note recordings directly to Claude, Cursor, or any MCP-compatible AI client. Search, retrieve, and work with your meeting transcripts and AI summaries hands-free.
What It Does
This MCP server gives your AI assistant direct access to your Plaud recordings:
- List recordings — See all your recordings with dates, durations, and transcript status
- Get transcripts — Pull full speaker-diarized transcripts for any recording
- Get AI summaries — Retrieve Plaud's AI-generated meeting summaries
- Search — Find recordings by title/keyword
- Get recent — Quickly pull your most recent recordings with full transcripts
Setup (5 minutes)
1. Install dependencies
git clone https://github.com/YOUR_USERNAME/plaud-mcp.git
cd plaud-mcp
pip install -r requirements.txt
2. Get your Plaud auth token
Plaud doesn't offer public API keys, so you'll grab your session token from the web app:
- Open web.plaud.ai in Chrome
- Sign in to your account
- Open DevTools:
Cmd+Option+J(Mac) orCtrl+Shift+J(Windows) - Click the Network tab
- Refresh the page
- Click any request to
api.plaud.ai - In the request headers, find
Authorization: Bearer eyJ... - Copy everything after
Bearer— that's your token
3. Configure your token
cp .env.example .env
Open .env and paste your token:
PLAUD_TOKEN=eyJhbGciOiJSUzI1NiIs...your_full_token_here
4. Connect to your AI client
Claude Code
claude mcp add plaud -- python /path/to/plaud-mcp/server.py
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"plaud": {
"command": "python",
"args": ["/path/to/plaud-mcp/server.py"],
"env": {
"PLAUD_TOKEN": "your_token_here"
}
}
}
}
Cursor / Windsurf / Other MCP Clients
Follow the same pattern — point to server.py as the command with your token as an env var.
5. Test it
Ask your AI: "List my recent Plaud recordings" or "What did I discuss in my last meeting?"
Token Refresh
Plaud tokens expire periodically. When you get auth errors, just repeat step 2 to grab a fresh token and update your .env file.
Available Tools
| Tool | Description |
|---|---|
plaud_verify_token |
Check if your auth token is valid |
plaud_list_recordings |
List recordings with pagination |
plaud_get_transcript |
Get full transcript + AI summary for a recording |
plaud_search_recordings |
Search recordings by title keyword |
plaud_get_recent |
Get recent recordings with full transcripts |
Privacy
- This runs 100% locally on your machine
- Your Plaud token never leaves your computer
- No data is sent to any third party
- Only you can access your recordings through this server
License
MIT — use it however you want.
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。