Observed arrival · 2026-08-28
Stealpage crawls the web, then lets agents search the receipts
An AI web-scraping API that turns sites into searchable Markdown corpora for MCP, JSON, and .md clients.
Field notes
Stealpage escalates fetching when ordinary HTTP is insufficient, then turns retrieved pages into Markdown documents and heading-based chunks. Its displayed corpus includes 1,562 embedded chunks, 15,346 internal links, and material in 89 languages, with categories ranging from documentation and articles to pricing pages. Search combines vector and full-text ranking and can filter by the page’s indexed type, while MCP, JSON, and Markdown interfaces are listed for downstream agent use.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue