ShadowCrawl

Our take

A single Rust binary combining web search, extraction, crawling, deep research, and stateful browser automation, with progressive fallbacks from native fetch through Chromium rendering to human-in-the-loop for CAPTCHAs. Its distinctive move is packaging Playwright-scale browser control into one steps-array omni-tool, cutting the turn count for multi-step flows. Note the registry still carries the old ShadowCrawl name — the project rebranded to CortexScout under the cortex-works org — and the 'Playwright killer' framing outruns what a small, young project has proven. For self-hosters who want token-efficient scraping and automation in one binary.

reviewed by hand · 2026-07-05

Something wrong or dead here? Report it

Verification record

last verified

yesterday

github stars

68

last commit

29 days ago

archived

no

license

MIT

in registry since

2026-02-18

github.com/DevsHero/ShadowCrawl