RagScrape vs Firecrawl
Great scraper. RagScrape is the pipeline after it.
| Feature | RagScrape | Firecrawl |
|---|---|---|
| URL → clean Markdown | ✅ | ✅ |
| Heading-aligned RAG chunks (fence-safe) | ✅ | ⚠️ generic splitting |
| Embeddings in the same call | ✅ 384-dim, Workers AI | ❌ separate vendor |
| Change detection (skip unchanged pages) | ✅ contentHash / ifNoneHash | ❌ re-embed everything |
| llms.txt generation for any site | ✅ POST /llms-txt | ❌ |
| Docs-site crawler → files + llms.txt | ✅ ≤100 pages | ✅ at scale |
| JavaScript/SPA rendering | ❌ (on the roadmap) | ✅ |
| MCP server included | ✅ npx rag-scrape mcp | ✅ |
| Free no-signup tool | ✅ /convert | ⚠️ trial credits |
| Paid entry | $9/mo (2,000 req) | see current pricing |
Verdict. Choose Firecrawl if you need JavaScript-rendered SPA scraping at scale today. Choose RagScrape if your pipeline starts after the fetch: chunks aligned to headings, vectors, change-aware re-indexing and llms.txt — from $9/mo, free tier, no card.
Try RagScrape free — 50 req/mo, no card →
Just convert one URL
Comparisons are honest to the best of our knowledge (1970-01-01) — competitors' features change; check their docs. RagScrape's known limits are listed, not hidden.