Miniflux (RSS Reader) MCP. Manage all your feeds, articles, and knowledge in one conversation.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Miniflux (RSS Reader) MCP Server lets your AI client manage and read all your RSS feeds directly inside Miniflux. Use tools like `list_entries` to pull article content, discover new sources with `discover_subscriptions`, and organize knowledge using categories—all without leaving your chat window.
What your AI agents can do
Create api key
Generates a new API key for your Miniflux account.
Create category
Creates a new organizational category in your feed library.
Create feed
Adds a specific RSS feed URL to your subscriptions.
Finds available RSS feeds from any URL or creates an entirely new subscription source using discover_subscriptions or create_feed.
Retrieves lists of entries, fetches the full text content for a specific article (fetch_entry_content), or checks unread counts with get_feed_counters.
Adds categories using create_category, bookmarks individual articles (toggle_entry_bookmark), or marks entire feeds/categories as read to manage your reading flow.
Allows you to bulk update entry statuses with update_entries_status, refresh all feed data, or export the whole library via export_opml.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 CompatibleWaiting for input…
Miniflux (RSS Reader): 46 Tools for Feed & Entry Management
These tools give your AI client direct access to every function in Miniflux, letting you manage feeds, read articles, and organize content programmatically.
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 Miniflux (RSS Reader) on Vinkius019e38c1create api key
Generates a new API key for your Miniflux account.
019e38c1create category
Creates a new organizational category in your feed library.
019e38c1create feed
Adds a specific RSS feed URL to your subscriptions.
019e38c1create user
Creates a new user account (Admin only).
019e38c1delete api key
Removes an existing API key from the system.
019e38c1delete category
Permanently deletes a category you created.
019e38c1delete feed
Removes a specific feed subscription entirely.
019e38c1delete user
Deletes a user account (Admin only).
019e38c1discover subscriptions
Searches and finds potential RSS feeds based on a given URL.
019e38c1export opml
Exports your entire feed collection into an OPML file for backup or migration.
019e38c1fetch entry content
Retrieves the full, original article content for a specific entry ID.
019e38c1flush history
Clears out local history data for clean testing or use.
019e38c1get entry
Fetches details about one single article entry by its ID.
019e38c1get feed counters
Gets the count of unread and read articles for any given feed.
019e38c1get feed icon
Fetches an icon image used to represent a particular feed.
019e38c1get feed
Retrieves metadata and status for a specific feed subscription.
019e38c1get icon
Fetches an icon image by its unique ID.
019e38c1get integrations status
Checks which third-party services are currently connected and enabled.
019e38c1get me
Retrieves basic information about the current logged-in user.
019e38c1get user
Fetches details for a specific user account (Admin only).
019e38c1get version
Returns the application's version and build information.
019e38c1healthcheck
Performs a basic database check to confirm server uptime.
019e38c1import entry
Allows manual import of an entry that wasn't sourced from a feed.
019e38c1import opml
Imports an entire collection of feeds and entries from an OPML file.
019e38c1list api keys
Lists all API keys associated with the account.
019e38c1list categories
Retrieves a list of all currently existing categories.
019e38c1list category entries
Gets every article entry contained within a specific category.
019e38c1list entries
Lists articles, applying filters like date or status to narrow down results.
019e38c1list feed entries
Shows all entries belonging only to one specified feed ID.
019e38c1list feeds
Retrieves a list of every active and inactive feed subscription.
019e38c1list users
Lists all user accounts on the system (Admin only).
019e38c1liveness
Liveness probe
019e38c1mark category as read
Marks every article in an entire category as read instantly.
019e38c1mark feed as read
Changes the status of all entries within a specific feed to 'read'.
019e38c1mark user as read
Marks all content associated with a user account as read.
019e38c1readiness
Readiness probe
019e38c1refresh all feeds
Triggers an automatic background refresh for every single subscribed feed.
019e38c1refresh category
Forces a data update across all feeds linked to one category.
019e38c1refresh feed
Synchronously updates the content of one specific feed right now.
019e38c1save entry
Sends a selected article entry to an external service for archiving or sharing.
019e38c1toggle entry bookmark
Marks or unmarks an individual article as a saved bookmark (starring it).
019e38c1update category
Modifies the name or settings of an existing category.
019e38c1update entries status
Bulk updates the status (read/unread) for multiple entries at once.
019e38c1update entry
Changes the title or body content of a specific saved article entry.
019e38c1update feed
Updates the settings or source URL for an existing feed subscription.
019e38c1update user
Update a user (Admin only)
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 Miniflux (RSS Reader), 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 Miniflux. 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 46 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Manually tracking your knowledge sources is a time sink.
Today, managing your reading list means opening tabs. You click into Site A for an article, copy the link; you jump to Site B and bookmark it in Notion. Later, when you try to summarize what you read, you have dozens of disorganized links and bookmarks spread across different platforms.
With this MCP server, you simply tell your agent: 'Gather all my unread articles from tech blogs.' The agent runs `list_entries` across every subscribed feed, pulls the clean titles, and gives you a single, actionable list. No tabs opened, no copy-pasting required.
Miniflux (RSS Reader) MCP Server: Get articles structured instantly.
The biggest time drain is the status check—you have to manually visit feeds just to see if something new dropped. You waste time figuring out what's read, what's unread, and where it belongs.
Now, you ask your agent: 'Which of my top 5 sources has content I haven't seen yet?' The server uses `get_feed_counters` and `list_entries` to run the check instantly. You get a clear status report in seconds.
What you can do with this MCP connector
Yo. This Miniflux MCP Server lets your AI client manage and read every damn RSS feed you've got, right inside your chat window. You don't gotta jump to another tab or open the whole web app just to check a few articles. You treat your entire knowledge base like one single source of truth using natural language prompts.
Getting Your Feeds Set Up and Keeping 'Em Current
You can start finding new sources by running discover_subscriptions on any URL, which searches for potential feeds you might want to track; or you can manually add a specific feed using create_feed. If you need to cut ties with an old source, delete_feed takes it out. To make sure your whole setup is solid, you've got list_feeds to see every single subscription—active and inactive—and you can even check the metadata for a specific feed using get_feed.
If you want to know how many articles you haven't read from any given source, get_feed_counters gives you those numbers.
Reading and Digging into Content
When it comes to content, this thing is deep. You can get a high-level list of articles using list_entries, letting you narrow down results by date or status until you find what you need. Need the full story? Use get_entry to grab details on one article ID, and then call fetch_entry_content to pull the entire original text, stripped clean.
You can also use list_feed_entries to see only articles from a single feed, or check out all entries tied to an organizational grouping with list_category_entries. If you just need general info on what user is logged in, get_me hands that over.
Organizing and Marking Stuff as Read
This isn't just a reader; it's a filing cabinet. You can create new organizational buckets using create_category, and then you'll have list_categories to see what you've got set up. When you find something important, use toggle_entry_bookmark to star it—it saves it for later. If an article belongs in a specific grouping, you can update that membership using update_category.
To keep your reading flow clean, you have status tools: you can mark everything in an entire category as read with mark_category_as_read, or change the status of every single piece from one feed instantly using mark_feed_as_read. You've got bulk control too; update_entries_status lets you change the read/unread state for multiple articles all at once.
Advanced Feed Management and Maintenance
Keep your data fresh by triggering background updates for every single feed with refresh_all_feeds, or if you only care about one source, use refresh_feed. If a whole category's worth of feeds needs an update, run refresh_category. You can also manually bring in content that wasn't sourced from a feed using import_entry, and for massive migrations, the system handles entire collection imports via import_opml or backs up your whole library into a file with export_opml.
To save a specific piece of reading material outside the app, you use save_entry to send it somewhere else. If you ever need to update the source URL or settings for an existing feed, update_feed handles that.
Admin and Utility Stuff
For system control, you can generate a new secret key using create_api_key, list every key attached to your account with list_api_keys, or delete old ones with delete_api_key. If you're managing users—and this is only for admins—you can create accounts via create_user, manage them through list_users and fetching details using get_user, and update their profile data with update_user; there are also tools to delete those user accounts completely.
Finally, the system keeps itself running; you can check its status with healthcheck, or if you're just testing something out, flush_history clears local history data.
019e38c2-2aca-73ae-8c9a-c58b06bb586b How Miniflux (RSS Reader) MCP Works
- 1 First, subscribe to this server and provide your Miniflux Instance URL and API Key.
- 2 Next, prompt your AI client with a task (e.g., 'List all my current feeds' or 'Find RSS for Google').
- 3 The agent runs the necessary tools (like
list_feedsordiscover_subscriptions) and returns the data to you for review.
The bottom line is, your AI client uses the specific Miniflux API endpoints we expose to perform actions on your behalf, without needing direct web access.
Who Is Miniflux (RSS Reader) MCP For?
This is for anyone whose job involves reading and synthesizing information from multiple online sources. If you're a researcher constantly bouncing between 10 different blogs, or a developer trying to keep up with tech releases across half a dozen sites, this saves hours of manual bookmarking and copy-pasting.
Aggregates sources from diverse academic journals and news outlets. They use list_entries to pull article titles and then fetch_entry_content for deep reading, all without leaving their terminal.
Gathers material for newsletters or social media posts. They rely on discover_subscriptions to find new sources and use create_category to group articles by topic before writing copy.
Stays current with technical blogs, GitHub releases, and major framework updates. They often run prompts like 'Show me all unread entries from my tech feeds' using get_feed_counters.
What Changes When You Connect
- Consolidate Data: Instead of manually checking 15 different RSS pages, you use
list_entriesorget_feed_counters. Your AI agent pulls the latest titles and unread counts from all sources at once. - Structured Organization: You don't lose important reads. Use
create_categoryto build a custom knowledge structure, then assign articles with simple commands that calllist_category_entries. - Deep Content Access: Don't just read headlines. If you need the full text of an article, run
fetch_entry_content. This gets the original source material so your agent can summarize or process it properly. - Workflow Efficiency: Running a prompt like 'Mark everything from TechCrunch as read and save the rest to my Dev category' combines multiple functions—
mark_feed_as_read,create_category, andlist_entries—into one step. - Maintenance Control: When you know your data is stale, use
refresh_all_feeds. It forces all sources to update in the background, ensuring your agent always reads current information.
Real-World Use Cases
The Research Dumpster Fire
A researcher needs to consolidate 40 articles from various academic feeds. Instead of opening tabs and copying links, they ask their agent: 'Get the titles for all my unread entries that mention AI.' The agent runs list_entries with filters, giving them a clean list they can process.
The Content Deadline Crunch
A curator needs to draft a newsletter on 'Web Development'. They ask the agent to find all articles in their 'Dev' category that haven't been read, and then save the top three using save_entry. This pulls content from multiple sources into one actionable list.
The Tech Update Sweep
A developer wants to know which feeds are active but haven't been checked in a while. They ask for the feed status, and the agent runs get_feed_counters on all sources, immediately highlighting any feeds with unread articles.
The Tradeoffs
Treating it like a simple RSS reader.
Just opening the Miniflux web interface and manually marking things. You miss the context of cross-source comparison or automated organization.
→
Use your agent to run list_entries across multiple feeds at once, then use update_entries_status to mark them all as read in one prompt.
Assuming data is up-to-date.
Seeing an article title but knowing it might be old. The agent uses cached/stale data and gives bad advice.
→
Before running any query, explicitly trigger refresh_all_feeds. This forces the server to grab fresh data from every source.
Trying to change content directly in chat.
Telling the agent 'Change this article.' The tool only allows viewing/organizing, not editing the original text.
→
If you need to edit a saved piece of content, use update_entry after fetching it with get_entry. This modifies your local copy, not the source.
When It Fits, When It Doesn't
Use this server if your job involves reading and synthesizing information from multiple, diverse online sources (blogs, journals, news outlets). You need a central point to read, organize, and manage that data flow. It's ideal for knowledge workers who spend time consuming content.
Don't use it if you only need basic RSS feed functionality—if your requirement is simply to view feeds in the browser without AI processing or complex categorization, stick to a standard feed reader app. Also, don't use it if you only manage single sources; this tool shines when running list_entries across multiple distinct feeds.
If you need advanced filtering and content flow control, rely on tools like list_entries, create_category, and update_entries_status. These are the core functions that make manual feed management obsolete.
Common Questions About Miniflux (RSS Reader) MCP
How do I add a new feed using `create_feed`? +
You give your agent the RSS URL, and it executes the create_feed tool. It adds that subscription to your Miniflux account immediately.
What is the difference between `list_entries` and `list_feed_entries`? +
list_entries searches across multiple feeds based on filters (like date). list_feed_entries only shows you articles from one specific, designated feed ID.
How do I bulk mark items as read using this server? +
You use the dedicated status tools. For example, running 'Mark all entries in my 'Work' category as read' triggers mark_category_as_read instantly.
Does `export_opml` save my data outside Miniflux? +
Yes. It downloads an OPML file containing your entire feed collection, giving you a backup or migration package of all your subscriptions and entries.
If I use `list_api_keys`, how can my AI agent check which API keys are active or if one needs deletion? +
The tool lists all existing API keys associated with your account. It doesn't confirm 'activity,' but it shows what keys the system recognizes. You must then use delete_api_key manually to revoke access for any key you no longer need.
When I call `fetch_entry_content`, am I getting just the raw article text, or does it include metadata too? +
It provides the original, full content of the entry. This means you get the complete body text and often associated media data from the source, not just the summary snippet available in a feed list.
How does `refresh_all_feeds` operate? Is it instant, or is it a background job? +
It runs as an asynchronous background refresh. This means your AI agent sends the command and gets confirmation that the process started; you'll need to wait for the feed data to update before querying the latest content.
What information does `get_me` provide about my user profile within Miniflux? +
This tool returns your current user details, including basic account metrics and identifiers. It helps your agent confirm which identity it's operating under when making changes like marking entries as read.
Can I find RSS feeds from a website URL automatically? +
Yes! Use the discover_subscriptions tool with any website URL. The agent will scan the site and return all available RSS/Atom feed links found.
How do I mark all articles in a specific feed as read? +
Simply use the mark_feed_as_read action providing the feed_id. This will instantly update the status of all entries within that feed to 'read'.
Can I search for specific keywords across all my articles? +
Yes. Use the list_entries tool and provide a keyword in the search parameter. You can also filter by status (read/unread) or starred items.
Multi-server workflows that include Miniflux (RSS Reader) MCP
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.