browser-use
Verified · 34 days agoControl a real Chrome browser to complete any task: fill forms, extract data, book flights.
Kloakt
Verified · 26 days agoCloaked headless browser for AI agents — stealth TLS, smart extraction, SPA fallback
Install
claude mcp add kloakt -- uvx kloakt
Our take
A real Rust/V8 headless browser — derived, with credit, from the popular Obscura engine — wrapped in a Python MCP server exposing 12 tools: markdown extraction with SPA fallback, persistent stateful sessions, accessibility snapshots, scraping, and crawling. The lightweight claims (30MB memory, fast loads) are inherited from upstream, and the added session-persistence work looks genuine. Caveats: the whole point is stealth TLS fingerprinting and anti-detection (mind the terms of whatever you scrape), screenshots still shell out to system Chrome, the curl-pipe installer fetches release binaries without checksum verification, and it's a weeks-old v0.1 from a single pseudonymous maintainer.
Verification record
- 26 days ago
- 1
- 39 days ago
- no
- Apache-2.0
- 2026-04-28
More in web-automation
browse all →Scrapling MCP Server
Verified · 27 days agoWeb scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
mcp-server-browser
Verified · 22 days agoMCP server for browser use access
playwright-mcp
Verified · 29 days agoPlaywright Tools for MCP
invisible_playwright_mcp
Verified · 11 days agoAI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.
AIHawk
Verified · 22 days agoAI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.