From URL to knowledge base in 3 steps
Paste your URL
Enter any website URL. We automatically discover every page via your sitemap or by following links — no manual setup.
AI reads every page
Our crawler visits up to 25 pages, extracts all the content, and Claude AI classifies each page into the right category.
Copy structured articles
Get clean, compressed articles grouped by type. Copy individual cards or export everything in one formatted document.
0
Pages crawled per scan
0
Page types auto-detected
0s
From URL to knowledge base
0
Crawl data stored after you leave
9 page types detected automatically
The AI reads every page and classifies each article into the right category — no manual tagging needed.
Homepage
Value props, hero copy, brand voice
Products
Features, specs, descriptions
Policies
Privacy, terms, cookie notices
Shipping
Carriers, timelines, costs
Returns
Refunds, exchanges, warranty
FAQs
Help articles, common questions
About
Company, team, mission
Contact
Support channels, locations
Services
What you offer, how it works
Everything you need, automatically structured
The crawler does the heavy lifting and Claude AI organises the result, so you get a clean, copy-ready knowledge base in one click.
- Full site crawlDiscovers pages via sitemap + link following. Up to 25 pages per scan.
- AI classificationClaude AI reads every page and assigns the right category automatically.
- Copy-ready outputEach article has a copy button. Export all as formatted markdown.
- Privacy firstNo crawl data stored. Results live only in your browser session.
Who uses this tool
Shopify store owners
Scan your store and instantly generate a knowledge base to train your AI customer support agent on your policies, products and FAQs.
Try with your storeSupport teams
Extract help centre content from any existing website and format it for your ticketing system, Notion, or AI chatbot.
Extract your contentAgencies & consultants
Quickly onboard a new client's brand information by scanning their website and generating a structured brief in seconds.
Scan a client siteFrequently asked questions
How many pages does it crawl?
Up to 25 pages per scan. We discover pages via your sitemap.xml first, then follow internal links. For large sites, we prioritise the most important pages.
Does it work on JavaScript-rendered sites (SPAs)?
For most SPAs, yes — we extract metadata, Open Graph tags, and JSON-LD structured data even when the HTML body is minimal. For best results on heavily JS-rendered sites, try entering a specific page URL like /about or /faq.
Is there a limit to how often I can use it?
Free users can scan 1 URL per day. Raisolve subscribers (any plan, including the 30-day free trial) get unlimited scans and can import directly into their store's live knowledge base.
Can I import this into a help centre or AI chatbot?
Yes. The "Export all" button copies a clean markdown document you can paste into Intercom, Zendesk, Notion, or any AI tool. If you use Raisolve, it imports directly into your AI agent's live knowledge base.
Does scanning a website store any data?
No data from the crawled website is stored after you close the page. Your scan results exist only in your browser session.
What page types can it detect?
Nine categories: Homepage, Products, Services, Policies, Shipping, Returns, FAQs, About, Contact, and a catch-all Other category. The AI classifies each article automatically.