rss-feeds-mcp
Fetch, filter, and search RSS blog feeds to monitor industry content, track competitor blogs, and discover trending topics. 8 tools for feed management, article search, and content monitoring.
README
RSS Feeds MCP Server
A Model Context Protocol (MCP) server that lets AI assistants like Claude Desktop, Cursor, Windsurf, and any MCP client fetch, filter, and search RSS feeds. Monitor blogs, track industry news, and research content — all through natural language.
Zero configuration. Install and start reading RSS feeds immediately. No API keys, no authentication.
What Can You Do?
Ask Claude questions like:
- "What are the latest SEO blog posts this week?"
- "Search all feeds for articles about AI marketing"
- "Show me the latest posts from HubSpot"
- "Add TechCrunch to my feeds"
- "What content marketing articles were published today?"
Quick Start
npx rss-feeds-mcp
Or install globally:
npm install -g rss-feeds-mcp
Ships with default feeds — customize your own at ~/.rss-mcp/feeds.json.
Claude Desktop Configuration
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"rss": {
"command": "npx",
"args": ["-y", "rss-feeds-mcp"]
}
}
}
Restart Claude Desktop. Done — RSS feeds are now available in Claude.
Use with Other MCP Clients
Works with any MCP-compatible client including Cursor, Windsurf, VS Code + Cline, and more:
npx rss-feeds-mcp
Available Tools (8 Tools)
| Tool | Description |
|---|---|
list_categories |
List all available feed categories (SEO, content, social, email, analytics, CRM, news) |
list_feeds |
List all configured RSS feeds with their categories |
add_feed |
Add a new RSS feed with name, URL, and category |
remove_feed |
Remove a feed by name |
fetch_blogs |
Fetch latest posts from all feeds with date range and limit |
fetch_by_category |
Fetch posts filtered by category |
fetch_from_source |
Fetch posts from a specific named source |
search_blogs |
Search across all feeds by keyword (matches title and summary) |
Examples
Example 1: Add a feed and fetch latest posts
User prompt: "Add the HubSpot blog RSS feed and show me the latest 5 posts"
Expected behavior:
- Calls
add_feedwith name "hubspot", URL "https://blog.hubspot.com/rss", and category "content" - Then calls
fetch_from_sourcewith name "hubspot" and limit 5 - Returns the 5 most recent HubSpot blog posts with titles, publication dates, summaries, and links
Example 2: Search across all feeds for a topic
User prompt: "Search all my feeds for articles about 'content marketing' from the last month"
Expected behavior:
- Calls
search_blogswith keyword "content marketing" and range "30d" - Searches all configured feeds for articles containing "content marketing" in their title or summary
- Returns matching articles sorted by date with title, source, publication date, and link
Example 3: Get posts from a specific category
User prompt: "Show me the latest SEO articles from this week"
Expected behavior:
- Calls
fetch_by_categorywith category "seo" and range "1w" - Fetches posts from all feeds tagged with the "seo" category
- Returns articles from SEO-focused blogs published in the last 7 days, sorted by date
Use Cases
- Content Research — Find trending topics and content ideas from industry blogs
- Competitive Monitoring — Track what competitors are publishing
- News Aggregation — Stay on top of industry news across multiple sources
- Content Curation — Collect and filter articles for newsletters or social media
- SEO Research — Monitor SEO blogs for algorithm updates and best practices
- Marketing Intelligence — Track marketing trends across channels
Feed Configuration
Feeds are stored at ~/.rss-mcp/feeds.json. On first run, a default config is created with sample feeds. Manage feeds dynamically using the add_feed and remove_feed tools, or edit the file directly:
{
"categories": ["seo", "content", "social", "email", "analytics", "crm", "news"],
"feeds": [
{ "name": "hubspot", "url": "https://blog.hubspot.com/rss", "category": "content" },
{ "name": "techcrunch", "url": "https://techcrunch.com/feed/", "category": "news" },
{ "name": "searchengineland", "url": "https://searchengineland.com/feed", "category": "seo" }
]
}
Requirements
- Node.js >= 18
- No API keys or authentication needed
Related
- Model Context Protocol — The open standard for AI-tool integration
- Claude Desktop — Anthropic's desktop AI assistant
- MCP Server Registry — Curated list of MCP servers
Privacy Policy
See PRIVACY.md for our complete privacy policy.
TL;DR: This extension runs locally on your machine. All feed fetching happens directly from your machine to public RSS endpoints. No data is collected, stored, or transmitted to any third party by this MCP server. No API keys or authentication required.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。