A knowledge base from websites and chat answers — from one link

Send a link — the scenario reads the site and adds it to the knowledge base; then ask questions in chat and it answers from what's gathered.

  • Builds a knowledge base from sent links
  • Reads the site and turns it into clean text
  • Won't add one site twice
  • Answers questions from what's gathered
WebhookSupabaseScrape company website with FirecrawlDefault Data LoaderEmbeddings OpenAIReranker Cohere
Connect scrapingBee and goScriptera

How it works

To create Webhook records from scrapingBee without manual input, use a ready-made scenario.

  1. Starts when: Receive company URL
  2. Then: Validate and normalize URL
  3. Then: Check for duplicate in Supabase
  4. Check: Skip if already ingested
  5. If yes: Return duplicate notice
  6. If no: Scrape company website with Firecrawl
  7. Then: Store embeddings in Supabase
  8. Starts when: Return ingestion result
  9. Starts when: Return URL validation error
  10. Then: Load scraped content
  11. Then: Generate OpenAI embeddings
  12. Starts when: Receive chat message
  13. Then: Answer query from enriched leads
  14. Then: OpenRouter LLM
  15. Then: Chat memory
  16. Then: Generate OpenAI embeddings1
  17. Then: Rerank results with Cohere
  18. Then: Retrieve documents from Supabase

You can launch this scrapingBee + Webhook integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios