Linkwarden MCP. Manage web archives and knowledge base from chat.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Linkwarden connects your private web archive directly to your AI agent. It lets you treat all your saved bookmarks, articles, and web pages like a structured knowledge base.
You can list collections, create new links, preserve full-page archives (even if the original site dies), and manage metadata—all by just talking to your agent.
What your AI agents can do
Archive link
Forces an update of a link's archive files, saving the current state of the webpage.
Bulk update links
Adds or changes tags and collections for many links at once.
Create collection
Makes a new, empty folder (collection) to group related bookmarks.
Create, read, update, and delete link records and the groups (collections) they belong to.
Generate or retrieve full-page archive files for a specific web link ID, ensuring content preservation.
Fetch structured data about your entire knowledge base, including dashboard statistics and public configurations.
Execute authentication flows like password resets or email verification needed to secure the instance.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 CompatibleWaiting for input…
Linkwarden MCP Server: 32 Tools for Knowledge Management
These tools let you manage every aspect of your private web archive—from creating links to retrieving decade-old content.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Linkwarden on Vinkius019e38b7archive link
Forces an update of a link's archive files, saving the current state of the webpage.
019e38b7bulk update links
Adds or changes tags and collections for many links at once.
019e38b7create collection
Makes a new, empty folder (collection) to group related bookmarks.
019e38b7create highlight
Adds or updates specific notes and highlights on an existing link's content.
019e38b7create link
Adds a brand new URL to your entire knowledge base.
019e38b7delete collection
Permanently removes an entire folder of links and their contents.
019e38b7delete highlight
Removes a set of highlights or notes you previously added to a link.
019e38b7auth forgot password
Sends a password reset email to initiate account recovery.
019e38b7get archive
Retrieves the preserved archive file for a given link ID, ensuring content access.
019e38b7get avatar
Fetches the user's profile picture URL or identifier.
019e38b7get collection
Retrieves all details about a specific collection by its unique ID.
019e38b7get config
Gets the public settings and runtime configuration for your Linkwarden instance.
019e38b7get dashboard v1
Retrieves general dashboard metrics about your saved content using version 1 data.
019e38b7get dashboard v2
Retrieves detailed and structured dashboard metrics using the newest (version 2) data format.
019e38b7get link highlights
Fetches all the highlights and notes associated with a specific link.
019e38b7get link
Pulls all stored details for a single link based on its ID.
019e38b7get logins
Retrieves current login configuration settings for the system.
019e38b7get migration
Pulls all user data, collections, and links into a format ready to be exported or moved elsewhere.
019e38b7get preserved token
Generates a temporary URL token that allows viewing archived content for a limited time.
019e38b7get public collection links
Gets a list of links from a collection that is marked as public.
019e38b7get public collection tags
Retrieves the tags applied to a specific, public-facing collection.
019e38b7import migration
Accepts and processes data meant for migrating or restoring the entire knowledge base.
019e38b7list collections
Retrieves a list of every collection currently stored in your instance.
019e38b7list rss
Shows which RSS feeds are subscribed to by the system.
019e38b7auth reset password
Allows you to change your password using a secure token.
019e38b7stream preserved view
Streams a live, archived view of a web page in its preserved format for viewing.
019e38b7update collection
Changes the name or other properties of an existing collection folder.
019e38b7update dashboard v2
Modifies the layout and structure of the dashboard using version 2 data rules.
019e38b7update link
Changes core details (like the description or tags) of an existing link record.
019e38b7upload archive for link
Uploads a local archive file to replace or supplement the archived version for a specific link.
019e38b7upload standalone archive
Uploads an archive file and automatically creates a new, linked record in your system.
019e38b7auth verify email
Confirms that a provided email address belongs to the account.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Linkwarden, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Linkwarden. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 32 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Bookmark managers don't preserve context or history.
Today, when you bookmark something valuable, all you get is a link. That's it. You open it weeks later and realize two things: first, the site might be offline, second, you have no idea why you saved it in the first place—was it for work, or just random reading? You end up jumping between tabs, searching through old bookmarks, and realizing that your 'organized' library is actually a pile of dead ends.
With this MCP server, you tell your agent to act. Instead of dumping links into a single folder, the agent lets you use `list_collections` and then `create_collection` for specific projects. You can attach notes using `create_highlight` right when you save it. The result? A knowledge base that remembers *why* you saved the link.
Using Linkwarden MCP Server: Archiving and Retrieving Content
Manually verifying if a site is still up, or trying to find an old version of an article, used to involve guessing. You'd hope the original server was stable enough for you to view it years later—a huge gamble. The manual process required checking uptime monitors and hoping your browser cache held out.
Now, simply ask the agent: 'Show me the archive for this link.' The system runs `get_archive`, pulls the preserved file, and streams a clean, stable version of that content directly into your chat. It's instant access to history.
What you can do with this MCP connector
You connect your private web archive directly to your agent, treating every saved bookmark, article, and webpage like a structured knowledge base you can talk to. The system gives you full control over link records and collections without ever touching a browser. You'll find that the core function of this server is managing links—you can create new bookmarks using create_link, or pull all stored details for any single bookmark ID with get_link.
If you need to adjust an existing entry, you use update_link to change descriptions or tags; and if you've got a bunch of changes for multiple links, bulk_update_links handles adding or changing tags and collections for them all at once. To organize your bookmarks into logical groups, the agent lets you see every folder with list_collections, create an empty folder using create_collection, or change that folder's name or properties by calling update_collection.
You can also delete a whole group of links permanently using delete_collection. Getting details on any specific grouping is easy; simply requesting the collection ID runs get_collection.
When it comes to preserving content, you don't just get a link—you get an archive. The system retrieves the full preserved file for a given link ID with get_archive, guaranteeing access even if the original site goes belly-up. If you want to make sure the data is current, you can force an update of a link's archive files by running archive_link.
You can also upload local backups; either upload_standalone_archive takes a file and automatically makes it a new record, or if you just need to supplement an existing entry, you use upload_archive_for_link. For quick access, the agent generates a temporary URL token via get_preserved_token, letting you view archived content for a limited time.
You can even stream a live, preserved view of the webpage itself using stream_preserved_view.
You don't just save links; you annotate them. The system lets you add or update specific notes and highlights on an existing page’s content by calling create_highlight. To see everything noted down for one link, run get_link_highlights, and if you need to remove old notes, delete_highlight clears out a set of highlights or notes you previously added.
For managing public sharing settings, you can pull all links from a collection that's marked as public using get_public_collection_links, or retrieve the tags applied to a specific public-facing folder with get_public_collection_tags. You can also see which RSS feeds are subscribed to by calling list_rss.
Need an overview of your library? The agent fetches structured data about all your saved content, giving you dashboard metrics using both the older format (get_dashboard_v1) and the newest, detailed structure with get_dashboard_v2. If you need to change how that dashboard looks or behaves, update_dashboard_v2 handles those rules. You can also get the public settings and runtime configuration for your entire Linkwarden instance via get_config, and if there are login details floating around, get_logins retrieves them.
Handling user accounts is straightforward. If you forget your password, the system sends a reset email using auth_forgot_password; then you can change it with a secure token by calling auth_reset_password. To confirm that an email address belongs to the account, run auth_verify_email. For moving or backing up your data, you pull all user info—including collections and links—into export format using get_migration, and when you're ready to restore it on a new instance, you process that data with import_migration.
019e38b7-db66-70c9-bd5d-9021bfb6d305 How Linkwarden MCP Works
- 1 Subscribe to the Linkwarden server and provide your specific Instance URL and Personal Access Token.
- 2 Connect this server through your preferred AI client (Claude, Cursor, etc.) using the MCP protocol.
- 3 Tell your agent what you need. For example: 'Show me all collections' or 'Archive this link.' The agent executes the necessary tool call against Linkwarden.
The bottom line is that you talk to your AI client, and it handles talking to Linkwarden for you.
Who Is Linkwarden MCP For?
Anyone who collects links or research material regularly. This server is built for the researcher who's tired of losing source material; the developer needing a clean way to document API endpoints; and the knowledge worker who needs one single place to manage all their reading material, without leaving their chat window.
They use this to instantly organize sources. Instead of downloading PDFs or keeping random bookmarks, they ask the agent to list collections and archive specific paper links.
They manage technical documentation links and code snippets directly from their IDE/agent chat. If a link is critical, they tell the agent to update its metadata.
They maintain a clean, organized library of web resources. They ask the agent to create new collections for projects or bulk-update tags across dozens of links at once.
What Changes When You Connect
- Never lose a source. By using
get_archiveorupload_standalone_archive, you ensure that even if the original website goes down, your agent can retrieve the full content history. - Centralize everything. You don't have to jump between bookmark managers and note apps. The agent handles link creation (
create_link) and collection organization (list_collections) all in one chat session. - Manage scale without friction. Use
bulk_update_linksorupdate_collectionto apply tags, change folders, or modify dozens of links at once, instead of clicking through them individually. - Keep your research flowing. The agent allows you to instantly retrieve dashboard insights (
get_dashboard_v2)—you see exactly how much knowledge you've stored without leaving the chat window. - Control metadata directly. Need to add a specific note or highlight? Use
create_highlightto attach context to a link, making it more than just a URL.
Real-World Use Cases
The Research Deep Dive
A PhD student saves dozens of articles over six months. Instead of having 50 different bookmarks in 10 folders, they ask the agent to list_collections and then use a prompt like: 'Get me all links from my 'Quantum Physics' collection that need an archive update.' The agent runs get_link for each one, calls archive_link, and confirms it. Result: A fully archived, clean set of source material.
The Technical Documentation Update
A developer finds a new API endpoint link. They don't just copy the URL; they ask the agent to create_link, then immediately run create_highlight with notes on its usage, and finally use update_collection to place it in 'V3 APIs'. The process is fast, traceable, and keeps documentation organized.
The Knowledge Audit
A content team needs to know what resources they saved last quarter. They ask the agent for a summary using get_dashboard_v2. The agent returns metrics showing 400 links across 15 collections, allowing them to immediately identify gaps or areas of under-archiving.
The Content Migration
A user needs to move their entire library off the current system. They instruct the agent to run get_migration. The server packages up all user info, links, and collections into a single structured data payload, ready for export.
The Tradeoffs
Manual Link Cleanup
A developer manually opens 50 bookmarks in Chrome, copies the URL of each one, pastes it into a spreadsheet, and then updates them individually. This is slow, error-prone, and requires dozens of browser tabs.
→
Use the agent to run create_link or bulk_update_links. Just tell your AI client: 'Create 5 new links from this list and put them all in the 'API Endpoints' collection.' The agent handles all 50 calls instantly.
Losing Source Context
A researcher bookmarks a useful article but forgets to write down why it was important, or what topic it related to. When they return later, the link is useless because there's no context.
→
When you get a link using get_link, immediately ask the agent to run create_highlight and add notes like 'Source for Q3 Report on Climate Change.' This makes the link functional again.
Archival Drift
You rely solely on the browser's cache, which might expire or fail if the original site structure changes. You assume the bookmark is enough.
→
Always confirm content preservation by asking the agent to run archive_link. This forces the system to grab a full-page snapshot and store it using the dedicated archive tools.
When It Fits, When It Doesn't
Use this server if your primary need is structured, long-term preservation of web content. You need more than just a link; you need the ability to retrieve the archived state of that link years later. This setup excels when managing large volumes (hundreds or thousands) of links across multiple projects and needs bulk operations like bulk_update_links.
Don't use it if all you need is a simple, quick place to jot down 5-10 URLs for immediate reference—a basic note app works better. If your main concern is just collaboration on live documents (like Google Docs), those dedicated tools are fine. You only need Linkwarden when the data attached to the link (the highlight, the collection structure, and the archive file) is as important as the URL itself.
Common Questions About Linkwarden MCP
How do I organize my links using Linkwarden MCP Server? +
You can use the agent to run list_collections first, and then tell it to run create_collection for a new project. After that, you just need to point the link into place.
Can I update multiple links at once with Linkwarden MCP Server? +
Yes. You don't have to do it one by one. Just ask your agent to run bulk_update_links and list the tags or collections you want to change across all selected IDs.
What is the difference between `get_link` and `get_archive`? +
get_link gives you the metadata (title, description, links) about a saved item. get_archive retrieves the actual preserved content file for that link ID.
How do I add notes to an existing bookmark using Linkwarden MCP Server? +
You call create_highlight. You provide the link ID and the text you want to save. This attaches context directly to the record, making it searchable.
What if I need to move my data out of Linkwarden? Should I use `get_migration`? +
Yes, that's right. The agent runs get_migration, which gathers all your user info, links, and collections into one structured payload ready for you to export.
I forgot my password; how do I use `auth_forgot_password` via the Linkwarden MCP Server? +
It initiates a password reset email. This tool sends instructions to your registered email address, allowing you to regain access without needing manual support intervention.
What is the purpose of `get_preserved_token`, and how does it secure my archived links? +
It creates a short-lived URL token for preserved content. This mechanism ensures that even if the original website goes offline, your AI client can stream the archived version using the generated link.
How do I verify the current system settings or public runtime details using `get_config`? +
get_config pulls the server's public runtime configuration. This lets your agent read key metadata, confirming the operational status and specific parameters of your Linkwarden instance.
How can I see a list of all my bookmark folders? +
You can use the list_collections tool. It will retrieve all your existing collections, allowing the AI to show you how your bookmarks are organized.
Is it possible to retrieve an archived version of a saved link? +
Yes! Use the get_archive tool with the specific Link ID. This will fetch the preserved archive file associated with that link.
Can I create a new collection directly through the AI? +
Absolutely. Use the create_collection tool by providing a unique ID and the collection details in the payload.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.