M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
R

Raindrop MCP Server

Community
Connect Raindrop.io to your AI assistant with natural language
How to install ↓ Compare GitHub source repository ↗
Category
Cloud Storage #12 of 15
Stars
★ 188 Very popular
Transport
stdio (local process)
Runtime
Node.js
Credentials
API key / credential required
License
MIT
Last commit
Tools
17

Static review · not runListed 2026-08-07

FMRS 52/100 · D · Weakest Reliability 8/20 — Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

  • Individuals and teams who rely on Raindrop.io for bookmark management.
  • Users who want to control their bookmarks via an AI assistant using natural language.

Not for

  • Users without a Raindrop.io account or who cannot provide an API token.
  • Users who require an official hosted solution (the official Streamable HTTP endpoint exists, but this server is self-hosted).

Required permissions

  • Requires a Raindrop.io API token, which grants read and write access to the user's collections, bookmarks, tags, highlights, and trash.

Risks and side effects

  • Requires managing an API token; leakage may lead to unauthorized modifications of bookmark data.
  • Some destructive tools (e.g., empty_trash, remove_duplicates) require confirmation but are irreversible.
  • Community-maintained, may be less stable than official servers.

Setup

Before you start

Runtime：Node.js

RAINDROP_ACCESS_TOKEN requiredsecret A Raindrop.io API access token, created in the integrations settings at https://app.raindrop.io/settings/integrations.
  1. Ensure you have a Raindrop.io account and obtain an API access token from https://app.raindrop.io/settings/integrations.
  2. Set the environment variable RAINDROP_ACCESS_TOKEN.
  3. Run using npx: npx -y @adeze/raindrop-mcp, or add the server to your MCP client configuration with the command and token. For Claude Desktop, you can download the .mcpb file and configure it.
claude_desktop_config.json 
{
  "servers": {
    "raindrop": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "@adeze/raindrop-mcp@latest"
      ],
      "env": {
        "RAINDROP_ACCESS_TOKEN": "YOUR_RAINDROP_ACCESS_TOKEN"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

Terminal
claude mcp add raindrop -e RAINDROP_ACCESS_TOKEN=YOUR_RAINDROP_ACCESS_TOKEN -- npx @adeze/raindrop-mcp@latest

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After setting RAINDROP_ACCESS_TOKEN and connecting, check that collection_list and diagnostics appear in the client's tool list; calling collection_list and seeing your Raindrop collections confirms the connection.

Troubleshooting

  1. Ensure RAINDROP_ACCESS_TOKEN is set correctly.
  2. Verify the token is valid; you can regenerate it from Raindrop.io settings.
  3. If using npx, ensure npx and Node are installed and network access to the npm registry is available.

Things to try

Once connected, you can ask your AI assistant things like:

  • Search my bookmarks for anything tagged rust
  • List all my collections and show the hierarchy tree
  • Audit my bookmark library for broken links and duplicates
  • Bulk move all bookmarks in this collection to my reading list

Tools 17

diagnostics read-only
Server diagnostic information and library health metrics.
collection_list read-only
List all collections as a flat list.
get_collection_tree read-only
Hierarchical view of collections with full breadcrumb paths.
collection_manage writes
Create, update, or delete collections.
bookmark_search read-only
Advanced search with filters, tags, and pagination.
bookmark_manage writes
Create, update, or delete bookmarks.
get_raindrop read-only
Fetch a single bookmark by ID.
list_raindrops read-only
List bookmarks for a collection with pagination.
Show 9 more tools
get_suggestions read-only
AI-powered organization advice (tags/collections) for a URL or bookmark.
suggest_tags read-only
Suggest relevant tags from bookmark metadata using AI-assisted analysis.
bulk_edit_raindrops writes
Bulk update, move, or remove bookmarks in a specific collection.
tag_manage writes
Rename, merge, or delete tags.
highlight_manage writes
Create, update, or delete highlights.
library_audit read-only
Scan library for broken links, duplicates, and untagged items.
empty_trash destructive
Permanently empty the trash (requires confirmation).
cleanup_collections destructive
Remove empty collections (requires confirmation).
remove_duplicates destructive
Find and remove duplicate bookmarks with safe confirmation flow.

Use cases

Add, organize, and search bookmarks using natural language commands.
Automatically organize collections and tags with AI suggestions.
Perform bulk cleanup of duplicate bookmarks, fix broken links, and manage trash.

Supported clients

Claude Desktop
Gemini CLI
Codex
Claude Code
GitHub Copilot

Listed from the project's documentation, not tested by this site.

Overview

Raindrop MCP Server is a community-developed server that connects Raindrop.io, a bookmark management service, to AI assistants. It provides tools to create, update, and delete collections and bookmarks, search bookmarks by tags, domain, type, date, manage tags and highlights, bulk edit, audit broken links and duplicates, and manage trash. It supports stdio transport and can be configured for use with various clients like Claude Desktop, Gemini CLI, and others. Requires a Raindrop.io account and API access token.

Similar servers

Compare with other MCP servers

Source revision 5b1e7f96cf82 Data synced 2026-10-12 Read the FMRS scoring method