Agent tooling
jevcrawl is an open-source web crawler and scraper for AI assistants, using a small decision model to score links and passages so agents receive exact source excerpts rather than entire sites.
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
Created by Rayen Mlayeh, jevcrawl offers crawl, scrape, and map tools that extract verbatim text with source URLs and headings. It relies on TypeSafe's paid Jev API (pinned to jev-1.13.0) to rate relevance and manage a local cache with spend caps. The project is an early-stage alpha release. It does not block requests to private or internal network addresses. Injected webpage instructions can still distort scoring, and extracted content may contain hostile material.
Connect jevcrawl to your coding assistant with a TypeSafe service key, then give it a question and a starting website. It follows promising links and returns quoted passages with their locations. For a single known page, scraping without a goal avoids model scoring. Keep it away from private networks and treat returned text as untrusted.