DeepCrawl

Each URL scraped to Markdown. Concurrency set in Settings tab.

Returns hierarchical link tree from the page.

Paste an apologetics site, hit go. Downloads every page on that domain as clean Markdown. Stays on that domain only — won't wander off to other sites. Save the whole thing as one big .md file.

Pages containing ANY of these keywords (case-insensitive) will be saved. Leave blank to save everything.

The Spider starts at your seed URL, follows every link it finds, reads each page, and checks for your keywords. Matching pages get saved. It's a real web crawler.

Exa uses neural search. Describe what you want in plain English. Click "Scrape Selected" to pull full markdown, or "Send to Spider" to deep-crawl the results.

Searches your local SearXNG at 192.168.1.177:5147. Click "Send to Spider" to deep-crawl the results.

Searches Semantic Scholar (200M+ papers) and arXiv. "Open Access only" filters to papers with free PDF downloads. Click any PDF link to download directly.

These words get added to each search. Helps focus results. Leave blank to search topics as-is.

Paste your list of topics. For each one, it searches the web, grabs the top results, and scrapes every page to markdown. 646 topics x 10 results = ~6,460 pages. It'll take a while but it'll get them all. You can stop anytime and save what you have.

Stored in your browser only (localStorage). Never sent anywhere except Exa's API.

Your local SearXNG instance. Must have JSON format enabled.

How many pages to scrape at once. 5 is safe, 10 is aggressive, 20 is max.