This server wraps RSS feed parsing and article extraction into two straightforward tools. Use fetch_feed_entries to pull the latest headlines from any RSS feed URL with configurable limits, or fetch_article_content to grab the full text of an article and convert it to Markdown. It handles the feed parsing and content extraction so you can ask Claude to monitor news sources, aggregate updates, or pull complete articles without hitting paywalls or dealing with HTML. Supports both stdio for local Claude Desktop setups and HTTP streaming for containerized deployments. Handy when you want to stay on top of multiple feeds or need clean article text for summarization and analysis without switching contexts.
Public tool metadata for what this MCP can expose to an agent.
fetch_feed_entriesFetch RSS feed entries from a given URL2 paramsFetch RSS feed entries from a given URL
urlstringlimitintegerfetch_article_contentFetch and extract article content from a URL, formatted as Markdown1 paramsFetch and extract article content from a URL, formatted as Markdown
urlstringAn MCP (Model Context Protocol) server for RSS feed aggregation and article content extraction. You can use it to subscribe to RSS feeds and get article lists, or extract the full content of an article from a URL and format it as Markdown.
English | 中文
You can use this MCP server in MCP-capable clients such as Claude Desktop and CherryStudio.
For Claude Desktop, add the following configuration under the "mcpServers" section in your claude_desktop_config.json file:
{
"mcpServers": {
"rss-reader": {
"command": "npx",
"args": [
"-y",
"rss-reader-mcp"
]
}
}
}
Basic RSS feed fetching
Can you fetch the latest 5 headlines from the BBC News RSS feed? URL: https://feeds.bbci.co.uk/news/rss.xml
Full article content extraction
Please extract the full content of this article and format it as Markdown: https://example.com/news/article-title
fetch_feed_entriesFetch RSS entries from a specified URL
Parameters:
url (required string): RSS feed URLlimit (optional number): Maximum number of entries to return (default 10, max 100)Returns: A JSON object containing feed metadata and a list of entries (including title, link, publication date, and summary)
fetch_article_contentExtract article content from a URL and format it as Markdown
Parameters:
url (required string): Article URLReturns: A JSON object containing the title, Markdown content, source URL, and timestamp
This server supports two transport modes:
Available environment variables:
stdio (default) or httpStream.TRANSPORT=httpStream, the listening port (default 8081).TRANSPORT=httpStream, the listening address (default localhost). In Docker, set this to 0.0.0.0 to expose the port externally.How to switch transport modes:
command + args configuration (see example above).TRANSPORT=httpStream and specify PORT (defaults to 8081 if not set).MCP_SERVER_HOST=0.0.0.0 and map the port.You can also run this MCP server in a Docker container. First, build the image in the project root:
docker build -t rss-reader-mcp .
Using CherryStudio as an example, the following configuration shows how to run this server over HTTP:
{
"mcpServers": {
"rss-reader-mcp": {
"command": "docker",
"args": [
"run",
"--rm",
"-p",
"8081:8081",
"-e",
"PORT=8081",
"rss-reader-mcp"
]
}
}
}
https://feeds.bbci.co.uk/news/rss.xmlhttps://techcrunch.com/feed/https://hnrss.org/frontpagehttps://www.technologyreview.com/feed/com.mcparmory/google-search
io.github.pipeworx-io/brave-search
marcopesani/mcp-server-serper
brave/brave-search-mcp-server
com.mcparmory/google-search-console
acamolese/google-search-console-mcp