Skip to main content

MCP server for Memos built with FastMCP - self-hosted note-taking tool

Project description

Memos MCP Server

PyPI Python License

MCP server for Memos - the open-source, self-hosted note-taking tool. Built with FastMCP.

Features

Tools

Memo CRUD:

  • create_memo - Create a new memo with content, visibility, and tags
  • get_memo - Get a memo by ID
  • update_memo - Update memo content, visibility, or pinned status
  • delete_memo - Delete a memo
  • list_memos - List memos with filters (state, custom filter)
  • search_memos - Search memos by content or tags

Tag Management:

  • list_tags - List all tags with usage counts
  • rename_tag - Rename a tag across all memos
  • delete_tag - Delete a tag from all memos

Visibility Control:

  • set_visibility - Set memo visibility (PRIVATE, PROTECTED, PUBLIC)
  • list_public_memos - List all public memos
  • list_private_memos - List all private memos

Resources

  • memos://list - Recent 20 memos summary (ID, title, tags, time, visibility)
  • memos://{memo_id} - Full details of a specific memo
  • memos://tags - All tags with usage counts
  • memos://stats - Statistics (total memos, by visibility/state, tags count, pinned)

Prompts

  • search_by_tag - Quick search for memos with a specific tag
  • create_quick_note - Create a simple memo with optional tag
  • find_recent_notes - Find memos from past N days
  • organize_by_visibility - Review and organize memos by visibility
  • update_memo_prompt - Update a specific memo's content with verification

Installation

# Using pip
pip install memos-fastmcp

# Using uvx (recommended)
uvx memos-fastmcp

Configuration

Get API Token

  1. Open your Memos instance
  2. Go to Settings > API Keys
  3. Generate a new API token

Environment Variables

MEMOS_URL=http://your-memos-instance:5230
MEMOS_TOKEN=your-api-token

Optional:

MEMOS_API_VERSION=v1          # Default: v1
MEMOS_REQUEST_TIMEOUT=30.0    # Default: 30.0 seconds
DISPLAY_STYLE=list            # Default: list (options: list, table)

Output Format

Query tools (list_memos, search_memos, get_memo, list_public_memos, list_private_memos) return Markdown format with two display styles:

Style Description
list Detailed list items with full timestamps and status indicators (default)
table Compact table format for quick scanning

Example list style:

## 📝 V2xmw6RkAEDo4TimKhCGNL

🔗 [查看详情](http://localhost:5230/memos/V2xmw6RkAEDo4TimKhCGNL)

**创建时间:** 2026-04-12 14:30:25
**状态:** 置顶 | 公开
**标签:** #work #project

项目计划已完成初步设计...

Example table style:

| ID | 创建时间 | 状态 | 标签 | 内容 |
|----|----------|------|------|------|
| V2xmw6RkAEDo4TimKhCGNL | 2026-04-12 | 置顶 \| 公开 | #work #project | 项目计划已完成... |

Notes:

  • ID column is not truncated (easy to copy for subsequent operations)
  • Content truncated: 200 chars in list style, 50 chars in table style
  • Hyperlinks point to browser-accessible memo detail pages

Usage with MCP Clients

Claude Desktop / Claude Code

Add to your MCP configuration:

{
  "mcpServers": {
    "memos": {
      "command": "uvx",
      "args": ["memos-fastmcp"],
      "env": {
        "MEMOS_URL": "http://your-memos-instance:5230",
        "MEMOS_TOKEN": "your-api-token"
      }
    }
  }
}

For Claude Desktop, place this in:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

For Claude Code, place this in .claude/settings.json in your project.

Other MCP Clients

# Run directly with environment variables
MEMOS_URL=http://your-memos-instance:5230 \
MEMOS_TOKEN=your-api-token \
uvx memos-fastmcp

Example Usage

Once configured, you can interact with Memos through Claude:

User: "Create a memo about my project plan with tag 'work'"
Claude: [Uses create_memo tool] "Created memo with ID: abc123..."

User: "Search for all memos tagged 'work'"
Claude: [Uses search_memos tool] "Found 5 memos tagged 'work'..."

User: "What are my most used tags?"
Claude: [Uses memos://tags resource] "Your top tags are: work (12), ideas (8)..."

User: "Show me recent memos from the past week"
Claude: [Uses find_recent_notes prompt] "Here are 7 memos from the past week..."

User: "Pin memo abc123"
Claude: [Uses update_memo tool] "Successfully pinned memo abc123"

Development

Local Setup

git clone https://github.com/Jiraiya8/memos-mcp.git
cd memos-mcp

# Install dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run locally
MEMOS_URL=http://localhost:5230 MEMOS_TOKEN=your-token python -m memos_mcp.server

Build & Publish

python -m build
twine upload dist/*

API Compatibility

Compatible with Memos API v1. Requires Memos 0.22.0 or later.

License

MIT License - see LICENSE for details.

Links

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

memos_fastmcp-0.2.1.tar.gz (58.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

memos_fastmcp-0.2.1-py3-none-any.whl (19.8 kB view details)

Uploaded Python 3

File details

Details for the file memos_fastmcp-0.2.1.tar.gz.

File metadata

  • Download URL: memos_fastmcp-0.2.1.tar.gz
  • Upload date:
  • Size: 58.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for memos_fastmcp-0.2.1.tar.gz
Algorithm Hash digest
SHA256 cd0e761e86dbbfaf5aa753c3870d9a140cb086027179cac3de8dddaef6382006
MD5 ea1e6a6a6b2a25b1d472e1a1eba59268
BLAKE2b-256 ae5a5cd881ca00ced55b58c1aecc8fd92744e45891c9be33b0fe91964052d145

See more details on using hashes here.

File details

Details for the file memos_fastmcp-0.2.1-py3-none-any.whl.

File metadata

  • Download URL: memos_fastmcp-0.2.1-py3-none-any.whl
  • Upload date:
  • Size: 19.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for memos_fastmcp-0.2.1-py3-none-any.whl
Algorithm Hash digest
SHA256 cd6f2bc8743758df4f920146ac6c5aa3d8b66d47b5e824f45b797a76539562d5
MD5 67ea4ff2d4ad928123de325bebfa34b8
BLAKE2b-256 20110452c4e91a4b52ebf7cbcb3cf7465c8fe0e344f9268f620d18d416e45fc8

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page