Mcp Server Scraper

What is Mcp Server Scraper?

Mcp Server Scraper extracts clean, readable markdown content from any URL using Mozilla Readability. It returns text, links, and metadata with no API keys or configuration, serving as a free alternative to Firecrawl for scraping documentation, blogs, and articles.

How to use Mcp Server Scraper?

Install and run with npx mcp-server-scraper, then configure it in your MCP client (e.g., Claude Desktop, Cursor, VS Code) by adding the command to the client’s mcpServers JSON. Once connected, use the provided tools like scrape_url to extract content.

Key features of Mcp Server Scraper

• scrape_url – extract clean text via Readability
• extract_links – get all links with href and anchor text
• extract_metadata – retrieve title, OG tags, favicon
• search_page – find a query string within a page
• scrape_multiple – batch scrape URLs for titles and excerpts
• No headless browser, no API keys, no accounts required

Use cases of Mcp Server Scraper

• Scrape API documentation for AI summarization
• Extract all links from a blog post for analysis
• Retrieve OG metadata from a news article
• Search a long documentation page for a specific term
• Batch scrape multiple URLs to compare summaries

FAQ from Mcp Server Scraper

How does Mcp Server Scraper compare to Firecrawl?

It is a free, open-source alternative that performs the same core task for static server-rendered pages without needing a paid API key.

What dependencies or runtime does Mcp Server Scraper require?

Only Node.js and npx – no additional accounts, API keys, or headless browser are needed.

What types of pages work best?

Server-rendered pages such as documentation sites, blog posts, and articles. JavaScript-heavy single-page applications (SPAs) are not supported.

Does Mcp Server Scraper store scraped data anywhere?

No, it fetches and processes content on demand; no data is persisted locally or sent to any external service.

What transport or authentication does Mcp Server Scraper use?

It runs locally via npx and communicates with MCP clients over stdio. No network transport or authentication is required.