MCP-RSS-Crawler

Name: MCP-RSS-Crawler
Author: mshk

@mshk

代码仓库

关于 MCP-RSS-Crawler

RSS Crawler MCP Server

基本信息

分类

浏览器自动化

运行时

node

传输方式

stdio

发布者

mshk

提交者

Masahiko Tachizono

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "rss-crawler": {
      "command": "/path/to/bun",
      "args": [
        "run",
        "/path/to/mcp-rss-crawler/apps/mcp-server/src/mcp-cli.ts"
      ],
      "cwd": "/path/to/mcp-rss-crawler",
      "env": {
        "PORT": "5556",
        "DB_DIR": "/path/to/mcp-rss-crawler",
        "FIRECRAWL_API_KEY": "fc-<YOUR_FIRECRAWL_API_KEY>"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is MCP-RSS-Crawler?

MCP-RSS-Crawler is an MCP (Message Chain Protocol) server that fetches RSS feeds, caches them in a SQLite database, and shares the latest articles with LLMs via the MCP protocol. It is designed for developers and users who want LLMs to access real-time or curated RSS feed content.

How to use MCP-RSS-Crawler?

Clone the repository, configure a claude_desktop_config.json with the path to Bun, the server script, and environment variables (PORT, DB_DIR, FIRECRAWL_API_KEY). Install dependencies with bun install, then start Claude Desktop. The MCP endpoint accepts POST requests with a JSON body containing a messages array.

Key features of MCP-RSS-Crawler

Fetches and caches RSS feeds using a SQLite database
Implements the MCP protocol for seamless LLM integration
Supports filtering feeds by category, source, or keywords
Provides API endpoints to add, update, and delete feeds
Supports fetching articles from Firecrawl

Use cases of MCP-RSS-Crawler

Ask an LLM for the latest news from your curated RSS feeds
Filter and retrieve articles based on specific keywords or sources
Automate feed management (add, update, delete) through API calls

FAQ from MCP-RSS-Crawler

What are the runtime dependencies?

Bun runtime and a Firecrawl API key are required. The server can run with Claude Desktop or any MCP‑compatible client.

How are the feeds stored and where does data live?

Feeds are cached in a local SQLite database. The database directory defaults to ~/.mcp-rss-crawler and can be changed via the DB_DIR environment variable.

What transport protocol does the server use?

The MCP protocol uses HTTP POST requests. The server listens on a configurable port (default 5556) and expects JSON payloads with a messages array.

How is authentication handled?

The server requires a Firecrawl API key set via the FIRECRAWL_API_KEY environment variable. No other authentication mechanisms are documented.

Are there any known limitations?

The README advises checking network settings and firewall for connection issues. Detailed logging can be enabled with DEBUG=mcp-rss:*.

浏览器自动化分类下的更多 MCP 服务器

Fastcrw

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary.

286