A semantic terminal browser that renders web pages as structured numbered elements for LLM agents via MCP, enabling agents to navigate, interact, and extract data from any website.
Ultra-lightweight headless browser for AI agents. Provides MCP tools for navigating URLs, extracting structured content, and building autonomous agent workflows.
A lightweight MCP server for parsing HTML, fetching URLs, rendering terminal-style screenshots, and executing JavaScript on static HTML without external dependencies.
Zero-dependency browser automation CLI with 40+ commands. Navigate, click, fill forms, take screenshots, capture network traffic, and more — all via Chrome DevTools Protocol. Built-in accessibility tree snapshots for AI agents (500x fewer tokens than screenshot-based approaches). Works with any Chromium browser already installed. 50KB total, no Playwright or Puppeteer needed.
Enables AI agents to fully control a terminal browser via MCP, navigating pages, filling forms, logging in, and managing cookies with deterministic numbered links and forms. Also provides an interactive human CLI.