RSS / Atom Reader MCP. Turns external news feeds into chat-ready text.

Listen up. This server connects your AI client directly to any public RSS or Atom feed out there—no matter where it is in the world. You're dealing with massive external XML news feeds, and we take all that messy code and translate it into clean, usable text right in the chat window.

Forget having to open dozens of browser tabs just to check what’s new; your agent handles the parsing for you.

When you use read_default_feed, you set up a standard source—a feed that's always relevant, like your company blog or a favorite industry journal. You point us at it once, and then you can just ask, "What’s new from the default feed?" The agent instantly reads and reports on whatever content is published there.

It keeps track of this routine monitoring for you.

If you need to check something specific—like a deep dive into one article stream or a niche industry update—you use read_feed. You just drop in the arbitrary RSS or Atom URL, and the agent runs with it. This tool is powerful because you get total control over what data gets extracted.

You can specify an item limit right there in the prompt; need the top five articles? Just tell it to check for 'top 5' results, keeping your output focused and tight.

The functionality goes way beyond just dumping raw text. When we run read_feed, the system doesn't just grab a block of characters; it actively parses and extracts structured data points from every article. That means you get specific headlines, confirmed publication dates, and the original source name—not some garbled mess of XML tags.

You see exactly what you're getting.

For instance, if you feed it a complex news aggregator link, it pulls out discrete articles, giving you structured data for each one, allowing your agent to process them individually. This capability means you can analyze patterns across multiple feeds without the AI having to deal with formatting logic or raw XML structure.

It just gives you the clean article summaries.

Think about what this saves you: You don't have to manually check ten different sources every morning. You set your default feed using read_default_feed, and when you prompt it, you get a consolidated summary of everything that happened across that source. For specific research moments, you use read_feed to target only the data you need—maybe just financial news from a particular exchange's Atom feed, limited to five entries.

The process is clean: provide the URL, set the parameters, and your agent returns structured summaries.

It’s all about turning chaotic external streams into actionable intelligence inside your chat window. You point it at the data source; we handle the parsing and extraction of headlines, dates, and body text for every item you request.