Awesome MCPBrowser Automation

hanzili/comet-mcp

⭐ 180 TypeScript repository created 2026-01-03

comet-mcp is a TypeScript Model Context Protocol server that connects Claude Code, or any other MCP client, to Perplexity Comet, an AI browser. Its argument is a division of labour: search APIs return static text and cannot interact with a page, while conventional browser automation servers make the same model that writes code also decide every click and keystroke, which floods the context window. Here the coding model sends a high-level goal, Comet works out the navigation on its own, and only the result comes back. Six tools implement this. comet_connect attaches to Comet and starts it automatically if it is not running. comet_ask submits a task and waits for the answer. comet_poll checks progress on long-running work. comet_stop cancels the current task. comet_screenshot captures the page. comet_mode switches Comet between its search, research, labs and learn modes. The transport chain is MCP over stdio into the server, then the Chrome DevTools Protocol on port 9222 into the Comet browser. Setup needs only an entry in .mcp.json or the Claude Code config that runs npx comet-mcp, plus a local Comet installation; a COMET_PATH environment variable covers non-standard install locations. Requirements are Node.js 18 or newer and macOS, Windows or WSL2. WSL2 needs mirrored networking enabled in .wslconfig so the Linux side can reach Comet running on Windows, and the server detects WSL and talks to Windows through PowerShell; a clear error is shown when mirrored networking is missing. The README also covers connection failures caused by a wrong install path or an occupied port 9222. Typical uses are agentic research across several sites and tasks behind login walls or dynamic content. MIT licensed.

https://github.com/hanzili/comet-mcp

browser-automationperplexitycometclaude-codecdpresearchtypescriptmcp-server

Also in Browser Automation

microsoft/playwright-mcp

Playwright MCP is an MCP server that enables LLMs to automate and interact with web browsers using structured accessibility snapshots for efficient and deterministic browser automation.

alibaba/page-agent

Page Agent is an in-page JavaScript GUI agent that enables natural language control of web interfaces, offering easy integration without browser extensions or headless browsers.

Skyvern-AI/skyvern

Skyvern automates browser-based workflows using large language models and computer vision, providing a robust and scalable web automation solution that integrates with the Model Context Protocol (MCP).

firecrawl/firecrawl-mcp-server

Firecrawl MCP Server is an MCP server implementation that adds advanced web scraping, crawling, and search capabilities to LLM clients like Cursor and Claude, enabling enhanced data extraction and research functionalities.

AgentDeskAI/browser-tools-mcp

BrowserTools MCP is a powerful browser monitoring and interaction tool that leverages the Model Context Protocol to enable AI-powered applications to capture and analyze browser data directly within MCP-compatible IDEs.

BrowserMCP/mcp

Browser MCP is an MCP server and Chrome extension that enables AI applications to automate and control your browser locally using your existing browser profile for privacy, speed, and stealth.

apify/actors-mcp-server

Apify MCP Server is an MCP protocol server implementation that enables AI assistants to interact with Apify Actors for data extraction, web scraping, and automation tasks through a standardized protocol.

agent-infra/sandbox

agent-infra/sandbox is an all-in-one sandbox environment combining Browser, Shell, File, MCP, and VSCode Server in a single Docker container, providing a unified, secure, and agent-ready platform for AI agent development and testing.