mcpField
L
Link Crypt

Link Crypt

Remote
getlinkcrypt.comProductivity0Updated 2w ago

About Link Crypt

Link Crypt is a hosted MCP server for people who want to search and manage saved links through AI assistants. Its bookmark library keeps page copies, enabling searches across page text as well as titles, with read-only access or approved changes.

Key features

  1. 01Search saved links and page text
  2. 02Save links and organize them with tags
  3. 03Generate reading digests
  4. 04Find and merge duplicate links
  5. 05Update or delete links and tags

Use cases

  1. 01Find an article saved earlier
  2. 02Save a link with a topic tag
  3. 03Review reading activity with a digest
  4. 04Clean up duplicate bookmarks

MCP support

Exposes Link Crypt library search and management tools to MCP clients. Users connect through OAuth 2.1 with PKCE and approve whether the assistant can make changes.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.getlinkcrypt.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

Link Crypt MCP server

Connect Claude, ChatGPT, Copilot and other AI assistants to Link Crypt, so you can search, save, and tidy your saved links by asking in plain English.

Link Crypt is a bookmark library that keeps a copy of every page you save, so search finds words in the page itself, not just its title.

Server URL: https://mcp.getlinkcrypt.com/mcp

This is a hosted (remote) server. There is nothing to install. You sign in to your Link Crypt account and approve the connection; no API keys to copy.

What you can ask

  • "What did I save this week?"
  • "Find the article I saved about sourdough starters."
  • "Save this link and tag it recipes."
  • "Which tags am I using most?"
  • "Do I have any duplicate links? Merge them."
  • "Write me a digest of what I've been reading this month."

Connect

Claude (Claude.ai, Desktop, mobile): Settings → Connectors → Add custom connector → https://mcp.getlinkcrypt.com/mcp, then sign in and approve.

ChatGPT: Settings → Apps & Connectors → Advanced settings → Developer mode on → Create → https://mcp.getlinkcrypt.com/mcp, Authentication: OAuth.

Claude Code

claude mcp add --transport http linkcrypt https://mcp.getlinkcrypt.com/mcp

VS Code (GitHub Copilot): .vscode/mcp.json

{
  "servers": {
    "linkcrypt": { "type": "http", "url": "https://mcp.getlinkcrypt.com/mcp" }
  }
}

Cursor: ~/.cursor/mcp.json

{
  "mcpServers": {
    "linkcrypt": { "url": "https://mcp.getlinkcrypt.com/mcp" }
  }
}

Full setup guide, including Grok, Hermes and Muse: https://getlinkcrypt.com/integrations/mcp

Tools

Read your library get_overview, search_links, list_links, get_link, list_tags, generate_digest, find_duplicates, search, fetch

Changes (turned off when you connect read-only) save_link, update_link, tag_link, update_tag, delete_link, delete_tag, merge_duplicates

Read tools are marked read-only. delete_link, delete_tag and merge_duplicates are marked destructive and need explicit confirmation.

Security and privacy

  • Sign-in is OAuth 2.1 with PKCE. You approve each connection on getlinkcrypt.com and choose whether the assistant may make changes. Disconnect any time from Preferences → Connected AI assistants, and access stops immediately.
  • A connection sees only your own library, with the same limits as your plan.
  • The server reads only what each tool is asked for. It does not see your chat history.
  • Privacy policy: https://getlinkcrypt.com/privacy

Support

[email protected]

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.