Skip to main content
Glama

linkding-mcp

⚠️ Warning: This is an experimental, unofficial MCP server in very early development stage. The code is "vibe coded", minimally tested and may be unstable. Not recommended for production environments.

A Model Context Protocol (MCP) server for linkding, a self-hosted bookmark manager. This is not an official Linkding package.

Features

  • 🔍 Bookmark Management: Create, update, delete and search bookmarks

  • 🏷️ Tag Management: Create, list and add tags to bookmarks

  • 📦 Archiving: Archive and unarchive bookmarks

  • 🔎 Advanced Search: Full-text search in bookmarks with various filters

  • 📊 Statistics: Overview of bookmark statistics

  • 📂 Bundle Management: Create and manage bookmark collections

  • 📎 Asset Management: Manage file assets (uploads and snapshots)

  • 👤 User Profiles: Access user profile and settings

  • 🕒 Advanced Filters: Filter by modification and creation date

  • 🔧 Configurable: Optional automatic scraping can be disabled

Related MCP server: Chrome Bookmark MCP Server

Installation

uvx linkding-mcp

Via pip

pip install linkding-mcp

Development installation

git clone https://github.com/jensneuhaus/linkding-mcp.git
cd linkding-mcp
pip install -e .

Configuration

Environment Variables

The server requires these environment variables:

  • LINKDING_URL: The URL to your linkding instance (e.g. https://bookmarks.example.com)

  • LINKDING_API_KEY: Your linkding API key (found under /settings/integrations)

MCP Client Configuration

Add the server to your MCP client configuration:

Claude Desktop (cline.json)

Direct from GitHub Repository (recommended):

{
  "mcpServers": {
    "linkding": {
      "command": "uvx",
      "args": [
        "--from", "git+https://github.com/jensneuhaus/linkding-mcp.git",
        "linkding-mcp"
      ],
      "env": {
        "LINKDING_URL": "https://...",
        "LINKDING_API_KEY": "..."
      }
    }
  }
}

Via PyPI (if available):

{
  "mcpServers": {
    "linkding": {
      "command": "uvx",
      "args": ["linkding-mcp"],
      "env": {
        "LINKDING_URL": "https://your-linkding.example.com",
        "LINKDING_API_KEY": "your-api-key-here"
      }
    }
  }
}

Using with direct Python

{
  "mcpServers": {
    "linkding": {
      "command": "python",
      "args": ["-m", "linkding_mcp.server"],
      "env": {
        "LINKDING_URL": "https://your-linkding.example.com", 
        "LINKDING_API_KEY": "your-api-key-here"
      }
    }
  }
}

Available Tools

📚 Reading List

  • get_reading_list - Get all bookmarks with #readlater tag

  • add_to_reading_list - Add bookmark directly to reading list

Bookmark Management

  • create_bookmark - Create new bookmark (with optional disable_scraping)

  • get_bookmarks - List bookmarks (with search, filter and date options)

  • get_bookmark - Get single bookmark by ID

  • update_bookmark - Update bookmark details

  • delete_bookmark - Delete bookmark

  • check_bookmark - Check if URL already exists as bookmark

Archiving

  • archive_bookmark - Archive bookmark

  • unarchive_bookmark - Remove bookmark from archive

Tag Management

  • get_tags - List all available tags

  • create_tag - Create new tag

  • add_tag_to_bookmark - Add tag to existing bookmark

Bundle Management

  • get_bundles - List all bundles

  • get_bundle - Get single bundle by ID

  • create_bundle - Create new bundle

  • update_bundle - Update bundle

  • delete_bundle - Delete bundle

Asset Management

  • get_bookmark_assets - List assets of a bookmark

  • get_bookmark_asset - Get single asset

  • delete_bookmark_asset - Delete asset

Advanced Features

  • search_bookmarks - Advanced full-text search with filters

  • get_bookmark_stats - Statistics about your bookmark collection

  • get_user_profile - Get user profile and settings

Examples

Creating a new bookmark

# Via MCP client - simple
create_bookmark(
    url="https://example.com",
    title="Example Website",
    description="An example website",
    tag_names=["example", "test"]
)

# With disabled scraping
create_bookmark(
    url="https://example.com",
    title="Manual Title",
    description="Manual Description",
    disable_scraping=True
)

Search and filter bookmarks

# All unread bookmarks
search_bookmarks(unread=True)

# Search by text
search_bookmarks(q="python tutorial")

# Filter by tags
get_bookmarks(q="tag:programming")

# Filter by date
get_bookmarks(added_since="2026-01-01T00:00:00Z")

# Filter by bundle
get_bookmarks(bundle=1)

Bundle management

# Create new bundle
create_bundle(
    name="Python Resources",
    search="python tutorial",
    any_tags="python programming",
    excluded_tags="outdated"
)

# List bundles
get_bundles()

# Update bundle
update_bundle(
    bundle_id=1,
    name="Updated Python Resources",
    order=5
)

Asset management

# Show assets of a bookmark
get_bookmark_assets(bookmark_id=123)

# Get asset details
get_bookmark_asset(bookmark_id=123, asset_id=456)

# Delete asset
delete_bookmark_asset(bookmark_id=123, asset_id=456)

Development

Setup

git clone https://github.com/jensneuhaus/linkding-mcp.git
cd linkding-mcp
pip install -e ".[dev]"

Running tests

pytest

Code formatting

black src tests
isort src tests
ruff check src tests

License

MIT License - see LICENSE for details.

Contributing

  1. Fork the repository

  2. Create a feature branch (git checkout -b feature/amazing-feature)

  3. Commit your changes (git commit -m 'Add amazing feature')

  4. Push to the branch (git push origin feature/amazing-feature)

  5. Open a Pull Request

API Reference

For more details about the linkding REST API, see the official documentation.

Available Tools

24 tools
add_tag_to_bookmarkB

Add a tag to an existing bookmark.

Inputs: tag_name (required), bookmark_id (required). Returns: result

add_to_reading_listA

Add a bookmark directly to your reading list (automatically tagged with readlater).

Inputs: url (required), notes, title, description. Returns: result

archive_bookmarkC

Archive a bookmark.

Inputs: bookmark_id (required). Returns: result

check_bookmarkB

Check if a URL is already bookmarked and get metadata.

Inputs: url (required). Returns: result

create_bookmarkC

Create a new bookmark in linkding.

Inputs: url (required), notes, title, shared, +5 more. Returns: result

create_bundleC

Create a new bundle in linkding.

Inputs: name (required), order, search, all_tags, +2 more. Returns: result

create_tagB

Create a new tag in linkding.

Inputs: name (required). Returns: result

delete_bookmarkB

Delete a bookmark.

Inputs: bookmark_id (required). Returns: result

delete_bookmark_assetC

Delete a bookmark asset.

Inputs: asset_id (required), bookmark_id (required). Returns: result

delete_bundleB

Delete a bundle.

Inputs: bundle_id (required). Returns: result

get_bookmarkA

Get a specific bookmark by ID.

Inputs: bookmark_id (required). Returns: result

get_bookmark_assetB

Get a specific bookmark asset by ID.

Inputs: asset_id (required), bookmark_id (required). Returns: result

get_bookmark_assetsB

Get list of assets for a bookmark.

Inputs: bookmark_id (required). Returns: result

get_bookmarksC

Get list of bookmarks from linkding.

Inputs: q, limit, bundle, offset, +3 more. Returns: result

get_bookmark_statsA

Get statistics about the bookmark collection.

Returns: JSON string with bookmark statistics (total, unread, archived, shared, tags)

Inputs: no named parameters. Returns: result

get_bundleA

Get a specific bundle by ID.

Inputs: bundle_id (required). Returns: result

get_bundlesC

Get list of bundles from linkding.

Inputs: limit, offset. Returns: result

get_reading_listA

Get all bookmarks from your reading list (tagged with #readlater).

Inputs: limit. Returns: result

get_tagsB

Get list of tags from linkding.

Inputs: limit, offset. Returns: result

get_user_profileA

Get user profile and preferences.

Returns: JSON string with user profile and preferences

Inputs: no named parameters. Returns: result

search_bookmarksC

Search bookmarks with advanced filters.

Inputs: q, limit, offset, shared, +1 more. Returns: result

unarchive_bookmarkA

Unarchive a bookmark.

Inputs: bookmark_id (required). Returns: result

update_bookmarkC

Update an existing bookmark.

Inputs: bookmark_id (required), url, notes, title, +5 more. Returns: result

update_bundleC

Update an existing bundle.

Inputs: bundle_id (required), name, order, search, +3 more. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.2/5.0

    Scored across 24 tools

    Disambiguation4/5

    Most tools are clearly distinct, but create_bookmark and add_to_reading_list overlap (the latter is just a tagged create), and get_reading_list could be replicated via search_bookmarks with a tag filter. Minor ambiguity exists but descriptions help.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun snake_case pattern (e.g., create_bookmark, get_bookmark_stats, delete_bundle). There is no mixing of conventions or vague verbs.

    Tool Count3/5

    With 24 tools, this is on the heavy side for a bookmark manager, spanning bookmarks, tags, bundles, assets, stats, and user profile. Each tool has a clear role, but the count feels slightly bloated and could be consolidated.

    Completeness4/5

    Bookmark lifecycle is well covered (CRUD, archive, search, reading list), but tags lack update/delete operations and there is no 'remove tag from bookmark' tool. Bundles and assets have appropriate coverage, though minor gaps remain.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      A Model Context Protocol server that provides persistent storage, categorization, and retrieval of bookmarks through AI clients like Claude. It allows users to manage web links using custom categories and includes options for local JSON storage or AWS S3 synchronization.
      2
      17
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      A Model Context Protocol (MCP) server implementation that provides seamless integration between Chrome bookmarks and AI assistants. This server enables AI models to access, search, analyze, and manage Chrome bookmarks through a standardized protocol.
      3
      MIT
    • A
      license
      A
      quality
      F
      maintenance
      A bookmark management MCP server that enables saving, searching, and managing bookmarks with tags and read/unread status, providing tools, resources, and prompts for integration.
      4
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server for Linkding bookmarks and web archival, enabling Claude to search, add, tag, archive, and delete bookmarks including archival snapshots.
      MIT