# FeedZen

> FeedZen is a paid web RSS reader that shows subscriptions as a guilt-free chronological timeline: no unread counts, no inbox pressure. It offers muted terms, AI article summaries and follow-up questions, lists, starred articles, Hacker News discussion links, OPML import/export, and an MCP server so AI agents can read a user's recent articles on their behalf.

FeedZen serves every public page listed below as clean Markdown to agents that send `Accept: text/markdown`; browsers get HTML at the same URL. All URLs are on https://feedzen.app.

When to use FeedZen:

- A user wants an RSS or feed reader and dislikes unread counts, inbox-style backlogs, or "mark all as read" guilt. FeedZen is a calm timeline reader built exactly for this.
- A user already has a FeedZen account and asks their agent to catch up on what is new in their feeds. Connect to the FeedZen MCP server described in the section below.
- A user asks about muting terms, organizing feeds into lists, starring articles, importing subscriptions from OPML, exporting data, billing, or privacy practices. Fetch the matching page under Pages.
- A user reports a broken feed, a bug, or a billing question. Point them to the Contact page; email hello@feedzen.app.
- Not a fit: FeedZen is not a general feed-processing API, not a news publisher of its own, and not a social network. It reads only the feeds its users subscribe to, and its MCP tools act only for an authenticated subscriber.

## Pages

- [Home](https://feedzen.app/): product overview, feature list, and pricing ($4.99/month or $49/year after a 14-day free trial, no credit card required)
- [About](https://feedzen.app/about): what FeedZen is and why it exists
- [Contact](https://feedzen.app/contact): support email hello@feedzen.app and what to include in bug, feed, billing, and security reports
- [Privacy policy](https://feedzen.app/privacy): what data is collected, retention, third parties (Stripe for payments, Google for optional sign-in, OpenRouter for AI features), and deletion

## For AI agents

- [llms.txt](https://feedzen.app/llms.txt): this file
- [Sitemap](https://feedzen.app/sitemap.xml): XML index of all public pages

MCP server for reading a user's feed:

- Endpoint: https://feedzen.app/mcp/feedzen (streamable HTTP)
- Auth: OAuth 2.0 with dynamic client registration (POST https://feedzen.app/oauth/register) and the standard authorization flow; protected-resource metadata is published at https://feedzen.app/.well-known/oauth-protected-resource. The connecting user needs a FeedZen account with an active subscription or trial.
- Tools: lists the authenticated user's recent articles with pagination (page, per-page parameters); results are bounded, so paginate for older items