mcp-notes
A Model Context Protocol (MCP) server implementation providing persistent note management created with Python SDK.
Documentation
MCP Notes Server
A Model Context Protocol (MCP) server implementation for managing notes with persistent storage.
Features
- Create, read, update, and delete notes
- Persistent storage using JSON
- Timestamp tracking for creation and modifications
- Note summarization via prompts
- Resource-based access using note:// URI scheme
Installation
Installing via Smithery
To install notes for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install notes --client claudeManual Installation
1. Ensure you have Python 3.10 or later installed
2. Create a virtual environment:
python -m venv .venv
# On Unix/MacOS:
source .venv/bin/activate
# On Windows:
.venv\Scripts\activate3. Install requirements:
pip install MCPProject Structure
notes/
├── __init__.py # Package initialization
├── server.py # Main server implementation
├── storage.py # Note persistence layer
├── resources.py # Resource handling (note:// URIs)
├── prompts.py # LLM prompt generation
└── tools/ # Server tools
├── __init__.py # Tools package initialization
├── list_tools.py # Tool listing functionality
└── handle_tools.py # Tool handling implementationAvailable Tools
- `add-note`: Create a new note
- `list-all-notes`: Display all stored notes
- `update-note`: Modify an existing note
- `delete-note`: Remove a note
Usage
1. Start the server:
mcp install src/notes
mcp start Notes2. Example operations:
# Create a note
await client.call_tool("add-note", {
"name": "example",
"content": "This is a test note"
})
# List all notes
await client.call_tool("list-all-notes")
# Update a note
await client.call_tool("update-note", {
"name": "example",
"content": "Updated content"
})
# Delete a note
await client.call_tool("delete-note", {
"name": "example"
})Storage
Notes are stored in `notes_storage.json` with the following structure:
{
"note_name": {
"content": "Note content",
"created_at": "2025-01-12T11:28:16.721704",
"modified_at": "2025-01-12T11:28:16.721704"
}
}Resource Access
Notes can be accessed as resources using the `note://` URI scheme:
- List resources: Returns all available notes as resources
- Read resource: Access a specific note using `note://internal/note_name`
Prompt Generation
The server includes a prompt generation feature for note summarization:
- Supports both brief and detailed summaries
- Formats notes for language model input
- Available via the "summarize-notes" prompt
Development
To modify or extend the server:
1. Clone the repository
2. Install development dependencies
3. Make changes in the appropriate module
4. Test thoroughly before deploying
Testing
Tests should cover:
- Basic CRUD operations
- Multiple note handling
- Error cases
- Resource access
- Prompt generation
License
[Add your license here]
Frequently asked questions
What is mcp-notes?
mcp-notes is A Model Context Protocol (MCP) server implementation providing persistent note management created with Python SDK.
How do I install mcp-notes?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is mcp-notes open source?
Yes — it is hosted on GitHub at https://github.com/truaxki/mcp-notes and has 7 stars.
Related MCP tools
🙌 OpenHands: Code Less, Make More for the Model Context Protocol. Enhance AI assistants with powerful integrations. Python-based implementation.
Universal memory layer for AI Agents; Announcing OpenMemory MCP - local and secure memory management. Python-based implementation.
基于大模型搭建的聊天机器人,同时支持 微信公众号、企业微信应用、飞书、钉钉 等接入,可选择ChatGPT/Claude/DeepSeek/文心一言/讯飞星火/通义千问/ Gemini/GLM-4/Kimi/LinkAI,能处理文本、语音和图片,访问操作系统和互联网,支持基于自有知识库进行定制企业智能客服。
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
An LLM agent that conducts deep research (local and web) on any given topic and generates a long report with citations. Built for the Model Context Protocol to
🚀 The fast, Pythonic way to build MCP servers and clients Trusted by 19900+ developers. Trusted by 19900+ developers. Trusted by 19900+ developers.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP