Blog RSS MCP Server

Blog RSS MCP Server

Enables interaction with blog RSS/Atom feeds to list posts, fetch content, get recent posts, retrieve blog metadata, and perform full-text search across post content. Supports any blog with an RSS/Atom feed through natural language queries.

Category
访问服务器

README

🌟Blog RSS MCP Server

A minimal Model Context Protocol (MCP) server that exposes tools for working with a blog’s RSS/Atom feed. It lists posts, fetches post content, returns recent posts, basic blog info, and can search post content on-demand.

What’s in here

  • main_server.py — MCP server with tools:
    • list_blog_posts() → [{title, slug (url), pubDate}]
    • get_blog_post(slug) → {slug, url, content} (content extracted from page HTML, preferring <article>)
    • get_recent_posts(count=5) → recent posts (simple reverse order)
    • get_blog_info() → blog metadata
    • search_full_text(query) → on-demand search (fetches each post and searches the text directly)
  • run_local_tests.py — quick local smoke test that exercises the tools without starting the MCP server.
  • Dockerfile — container image for running the server.
  • requirements.txt — Python dependencies.

Requirements

  • Python 3.10+
  • Recommended: a virtual environment (e.g., .venv)

Install dependencies:

pip install -r requirements.txt

Note for Python 3.13+: feedparser imports the removed stdlib module cgi; this repo includes python-legacy-cgi in requirements to ensure compatibility.

Configuration

Set environment variables as needed:

  • RSS_FEED_URL (required) → Full URL to your RSS/Atom feed (e.g., https://example.com/feed.xml)
  • MCP_TRANSPORT (optional) → stdio (default), sse, or streamable-http
  • MCP_MOUNT_PATH (optional) → mount path for applicable transports

Example:

export RSS_FEED_URL="https://www.sirishgurung.com/rss.xml"
export MCP_TRANSPORT=stdio

Run locally (MCP server)

python main_server.py

With a venv:

.venv/bin/python main_server.py

Local smoke test (no MCP client required)

.venv/bin/python run_local_tests.py

That script will print blog info, list posts, fetch one post’s content, and run a simple on-demand search.

Docker

Build the image:

docker build -t blog-rss-mcp:latest .

Run with stdio transport (for MCP clients that spawn the container and talk over stdio):

docker run --rm -i \
	-e RSS_FEED_URL="https://www.sirishgurung.com/rss.xml" \
	-e MCP_TRANSPORT=stdio \
	blog-rss-mcp:latest

Run with basic HTTP transport (optional):

docker run -d \
	-e RSS_FEED_URL="https://www.sirishgurung.com/rss.xml" \
	-e MCP_TRANSPORT=streamable-http \
	-e MCP_MOUNT_PATH="/mcp" \
	-p 5000:5000 \
	--name blog-rss-mcp \
	blog-rss-mcp:latest

Docker Desktop Setup With Claude Client

Prerequisites: Docker Desktop installed with MCP Toolkit enabled. Claude desktop installed.

  1. Build a Docker Image: You first build a local Docker image named blog-mcp-server using the provided project files (Dockerfile, Python server, etc.).
docker build -t blog-rss-mcp:latest .
  1. Define the Tool: You create a ~/.docker/mcp/catalogs/custom.yaml file that acts as a "catalog." This file defines the new "blog" tool, lists all its functions (like list_blog_posts, get_blog_posts, and others), and tells the system to use the Docker image you just built.
yamlversion: 2
name: blog-mcp-server
displayName: Blog RSS MCP Server
registry:
  blog-rss-mcp:
    description: "Local MCP server exposing blog RSS tools (list_blog_posts, get_blog_post, get_recent_posts, get_blog_info, search_full_text)."
    title: "Blog RSS MCP Server"
    type: server
    dateAdded: "2025-10-23T00:00:00Z"
    image: blog-mcp-server:latest
    ref: ""
    readme: "README.md"
    toolsUrl: ""
    source: ""
    upstream: ""
    icon: ""
    tools:
      - name: list_blog_posts
      - name: get_blog_post
      - name: get_recent_posts
      - name: get_blog_info
      - name: search_full_text
    metadata:
      category: productivity
      tags:
        - blog
        - rss
        - mcp
        - search
      owner: local
    env:
      - name: RSS_FEED_URL
        description: "URL to the blog RSS feed (required)."
        required: true
      - name: MCP_TRANSPORT
        description: "Transport used by FastMCP. Use 'stdio' for local client-managed lifecycle."
        default: "stdio"
      - name: MCP_MOUNT_PATH
        description: "Optional mount path for transports like SSE."
        default: ""

  1. Register the Tool: You edit the ~/.docker/mcp/registry.yaml file to officially register the new blog tool with the MCP (Model-Component-Processor) system.
registry:
  blog-rss-mcp:
    ref: ""
  1. Configure Claude Desktop: You edit the main claude_desktop_config.json file to tell the Claude application to load your new custom.yaml catalog file.
    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
    • Linux: ~/.config/Claude/claude_desktop_config.json

You can also find an Edit Config option in the Settings->Developers window. This step requires adding the file path to the command arguments and ensuring your home directory path is correct.

{
    "mcpServers": {
      "mcp-toolkit-gateway": {
        "command": "docker",
        "args": [
          "run",
          "-i",
          "--rm",
          "-v", "/var/run/docker.sock:/var/run/docker.sock",
          "-v", "[YOUR_HOME]/.docker/mcp:/mcp",
          "docker/mcp-gateway",
          "--catalog=/mcp/catalogs/docker-mcp.yaml",
          "--catalog=/mcp/catalogs/custom.yaml",
          "--config=/mcp/config.yaml",
          "--registry=/mcp/registry.yaml",
          "--tools-config=/mcp/tools.yaml",
          "--transport=stdio"
        ]
      }
    }
  }
  1. Restart and Test: After restarting the Claude Desktop app, the new blog functions will be available. You can test them by asking Claude to "get recent blogs" or "list all blogs"

Output Samples

"Screenshot of claude desktop using the mcp server"

Notes & troubleshooting

  • You can add your rss link to RSS_FEED_URL in the dockerfile to make it the default.
  • Logs go to stderr; stdio transport messages go to stdout (safe for MCP clients).
  • If HTTPS feed fetches fail in a minimal base image, install system CA certificates in your container.
  • If mcp.run() signature changes between versions, inspect FastMCP.run in your installed package and adjust the call accordingly.

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选