Anybox MCP Server

Anybox MCP Server

Enables interaction with Anybox bookmark manager on macOS, allowing users to search, save, and organize bookmarks with tags and folders through natural language.

Category
Visit Server

README

Anybox MCP Server

A Model Context Protocol (MCP) server for Anybox, the bookmark manager for macOS. This server allows Claude Desktop and other MCP clients to interact with your Anybox bookmarks.

Features

  • 🔍 Search bookmarks by keyword, tag, folder, or starred status
  • 🏷️ List all tags with IDs, names, and bookmark counts
  • 📁 List all folders with IDs, names, and bookmark counts
  • Save new bookmarks with tags, folders, comments, and starred status
  • 🔗 Direct integration with Claude Desktop for seamless bookmark management

Prerequisites

  • macOS with Anybox installed and running
  • Python 3.10 or higher
  • Anybox API key (see below)

Getting Your Anybox API Key

  1. Open Anybox
  2. Go to Preferences → General (or Settings → General)
  3. Copy the API key displayed there

Installation

  1. Clone or download this repository
cd ~/coding/anyboxTidier
  1. Create a virtual environment and install dependencies
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
  1. Set your API key

You'll set the API key in the Claude Desktop configuration (see below). No need to export it globally.

Usage with Claude Desktop

Configuration

Option 1: Automated Setup (Recommended)

Run the setup script which will guide you through the configuration:

./setup_claude_desktop.sh

This will:

  • Test your Anybox API connection
  • Automatically configure Claude Desktop
  • Verify everything is working

Option 2: Manual Configuration

  1. Locate your Claude Desktop config file:

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  2. Add the MCP server configuration:

{
  "mcpServers": {
    "anybox": {
      "command": "/Users/tommertron/coding/anyboxTidier/venv/bin/python",
      "args": [
        "/Users/tommertron/coding/anyboxTidier/anybox_mcp_server.py"
      ],
      "env": {
        "ANYBOX_API_KEY": "your-api-key-here"
      }
    }
  }
}

Important:

  • Replace /Users/tommertron/coding/anyboxTidier with the actual path to where you cloned this repo
  • Replace your-api-key-here with your actual Anybox API key
  • Make sure to use the full path to the Python interpreter in the venv/bin directory
  1. Restart Claude Desktop

Available Tools

Once configured, Claude will have access to these tools:

search_bookmarks

Search for bookmarks with various filters.

Parameters:

  • query (optional): Search keyword
  • tag_id (optional): Filter by tag ID
  • folder_id (optional): Filter by folder ID
  • starred (optional): "yes" or "no"
  • limit (optional): Max results (default: 50)

Example prompts:

  • "Search my Anybox for bookmarks about Python"
  • "Find all starred bookmarks"
  • "Search for bookmarks in the 'Work' folder"

list_tags

Get all tags with their IDs and bookmark counts.

Example prompts:

  • "List all my Anybox tags"
  • "What tags do I have?"

list_folders

Get all folders with their IDs and bookmark counts.

Example prompts:

  • "List all my Anybox folders"
  • "What folders do I have in Anybox?"

save_bookmark

Save a new bookmark with optional metadata.

Parameters:

  • url (required): URL or note content
  • comment (optional): Comment about the bookmark
  • tag_ids (optional): Array of tag IDs
  • folder_id (optional): Folder ID
  • starred (optional): Boolean (default: false)

Example prompts:

  • "Save https://example.com to Anybox"
  • "Save this URL to Anybox with the 'Programming' tag"
  • "Add https://github.com to my 'Work' folder in Anybox"

Example Workflow

Here's a typical conversation with Claude Desktop:

You: "List all my Anybox tags"
Claude: [Uses list_tags tool]
        "You have 15 tags. Here are some: Programming, Design, Research..."

You: "Search for bookmarks tagged with Programming"
Claude: [Uses list_tags to get the tag ID, then search_bookmarks]
        "Found 23 bookmarks tagged with Programming:
         1. Python Documentation - https://docs.python.org
         2. ..."

You: "Save https://github.com/anthropics/anthropic-sdk-python to Anybox
      with the Programming tag"
Claude: [Uses list_tags to get tag ID, then save_bookmark]
        "✅ Bookmark saved successfully!"

Troubleshooting

Connection Errors

Error: Connection refused or Error searching bookmarks

Solutions:

  1. Ensure Anybox is running
  2. Verify the API runs on http://127.0.0.1:6391
  3. Check your API key is correct
  4. Test the connection manually:
    curl -H "x-api-key: YOUR_KEY" http://127.0.0.1:6391/tags
    

API Key Issues

Error: 401 Unauthorized or Incorrect API Key

Solutions:

  1. Copy a fresh API key from Anybox Preferences
  2. Ensure the environment variable is set correctly
  3. Restart Claude Desktop after updating the config

Python/MCP Issues

Error: mcp module not found

Solution:

pip install mcp httpx

Error: Python version incompatible

Solution: Ensure Python 3.10+ is installed:

python3 --version

API Reference

This server uses Anybox's local HTTP API running on http://127.0.0.1:6391.

Discovered Endpoints

  • GET /search?q=&tag=&folder=&starred=&limit= - Search bookmarks
  • GET /tags - List all tags
  • GET /folders - List all folders
  • POST /save - Save a bookmark with {note, comment, tags[], folder, starred}

Authentication

All requests use the x-api-key header with your API key.

Development

Running the Server Directly

For testing:

python3 anybox_mcp_server.py

The server communicates via stdio using the MCP protocol.

Adding New Tools

To add new functionality:

  1. Add the tool definition in list_tools()
  2. Implement the handler in call_tool()
  3. Add any new API methods to the AnyboxAPI class

Credits

License

MIT License - See claude.md for Anybox API documentation sources.

Contributing

Contributions are welcome! If you discover additional Anybox API endpoints or features, please submit a PR.


Note: This is an unofficial integration. Anybox does not provide official public API documentation. This server is based on reverse-engineering from the Raycast extension and other community sources.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured