Install RAG Documentation Scraper in Claude Code
claude mcp add rag-documentation-scraper -e APIFY_TOKEN=... -- npx -y rag-docs-scraper-mcp-server{
"mcpServers": {
"rag-documentation-scraper": {
"command": "npx",
"args": [
"-y",
"rag-docs-scraper-mcp-server"
],
"env": {
"APIFY_TOKEN": "..."
}
}
}
}Replace each ... with your own value.
Frequently asked questions
What does the RAG Documentation Scraper MCP server do?
Crawl any documentation site and return heading-bounded, LLM-ready chunks with source metadata.
Is the RAG Documentation Scraper MCP server free?
The server is free, but you will likely need an account. The server itself is free, but it connects to a service that needs an account or API key, which may have paid tiers. Check the provider for current pricing.
Does RAG Documentation Scraper run locally or remotely?
Local. Runs on your machine (stdio). You install it with npx, uvx, Docker or from source.
How do I add RAG Documentation Scraper to Claude Code?
Run: claude mcp add rag-documentation-scraper -e APIFY_TOKEN=... -- npx -y rag-docs-scraper-mcp-server
More search & web scraping MCP servers
View allReplyNodes
replynodes/replynodes-mcp
Web MCP: scrape/crawl sites, web search, brand assets, app stores, YouTube, Reddit, Hacker News.
Public Official Asset Disclosure (공직자 재산공개)
yujiseok/public-official-mcp
Korean public official asset disclosure data (공직자 재산공개) - search and retrieve assets
Research
ilo415/research-mcp
AI research for agents. Company intel, competitor analysis, web scraping, topic dives. Pay via MPP.
Pixelvault
rsweeting1/pixelvault-mcp
AI-native marketplace for pre-cleared licensed media assets — search, license, and stream rights.
Rootz Archive
rootz-global/rootz-archive
Search your local Claude Code archive. For continuous capture, use the Rootz Archive plugin.
Openlibrary
cyanheads/openlibrary-mcp-server
Search books and authors, fetch editions, browse subjects, and resolve cover images.