ClassNotes MCP
Enables Claude Desktop to manage an Obsidian vault and automatically convert class recordings into structured study notes using local AI tools.
README
ClassNotes MCP
ClassNotes MCP is a personal Model Context Protocol (MCP) server built in Python that connects Claude Desktop with Obsidian to automate class note management.
The project allows Claude to interact with a local Obsidian vault through MCP tools, enabling it to create, read, list, and search notes.
It can also process class recordings automatically using:
- Faster-Whisper for local speech-to-text transcription.
- Qwen3:8B via Ollama to convert transcripts into structured study notes.
- Obsidian to store the generated notes as Markdown files.
- Claude Desktop as the MCP host.
Features
- Create notes directly in Obsidian.
- List notes by course.
- Read existing notes.
- Search concepts across the Obsidian vault.
- Convert class transcripts into structured notes.
- Process audio recordings and generate study notes automatically.
- Generate sections such as summaries, key concepts, review points, and Active Recall questions.
Workflow
Class recording
↓
Faster-Whisper
↓
Transcription
↓
Qwen3:8B (Ollama)
↓
Structured notes
↓
ClassNotes MCP
↓
Obsidian
Claude Desktop communicates with the Python server through MCP and can invoke the available tools according to the user's request.
Main Technologies
- Python 3.12
- Model Context Protocol (MCP)
- Claude Desktop
- Obsidian
- Faster-Whisper
- Ollama
- Qwen3:8B
Installation
Create and activate a virtual environment:
py -3.12 -m venv .venv
.\.venv\Scripts\Activate.ps1
Install the dependencies:
pip install -r requirements.txt
Create a .env file with the path to your Obsidian vault:
OBSIDIAN_VAULT="C:/path/to/your/Obsidian Vault"
Make sure Ollama is running and that qwen3:8b is available:
ollama list
Run
The MCP server can be tested with:
python server.py
or through MCP Inspector:
mcp dev server.py
Purpose
The goal of ClassNotes MCP is to reduce the time required to manually transcribe, summarize, and organize class recordings, while keeping the workflow local and using free/open-source tools whenever possible.
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。