io.github.ArkNill/browsegrabweb-automation

Browsegrab

Verified · 28 days ago

Token-efficient browser agent for local LLMs. Playwright + accessibility tree + MarkGrab.

Install

claude mcp add browsegrab -- uvx browsegrab

Our take

Implements a browser automation stack tuned for token efficiency, combining Playwright with accessibility-tree and page-grab techniques for local-LLM use. It is for developers building automated browsing agents, and the technical direction is plausible, though browser-agent projects tend to be brittle at the edges and require hands-on evaluation before relying on them heavily.

reviewed by hand · 2026-07-20

Something wrong or dead here? Report it

Verification record

last verified
28 days ago
github stars
0
last commit
6 mo ago
archived
no
license
MIT
in registry since
2026-03-18

github.com/QuartzUnit/browsegrab

More in web-automation

browse all →
browser-useweb-automation

browser-use

Verified · 34 days ago

Control a real Chrome browser to complete any task: fill forms, extract data, book flights.

hand-reviewed112k starschecked 34 days ago

Scraplingweb-automation

Scrapling MCP Server

Verified · 27 days ago

Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.

hand-reviewed79k starschecked 27 days ago

playwright-mcpweb-automation

playwright-mcp

Verified · 29 days ago

Playwright Tools for MCP

hand-reviewed37k stars5.9m dl/wkchecked 29 days ago

invisible-playwright-mcpweb-automation

invisible_playwright_mcp

Verified · 11 days ago

AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.

hand-reviewed32k starschecked 11 days ago

aihawkweb-automation

AIHawk

Verified · 22 days ago

AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.

hand-reviewed30k starschecked 22 days ago