Mcp Browser Agent
@yubao2000
Mcp Browser Agent について
让 AI Agent（Claude/ChatGPT/Codex）像人一样操作浏览器。36个工具：截图、点击、填表、提取、PDF导出、签页管理、Cookie持久化、批量自动化。开源免费，npm 全局安装即用。
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"browser-agent": {
"command": "mcp-browse-agent",
"args": []
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is Mcp Browser Agent?
A Model Context Protocol (MCP) server that enables AI agents (such as Claude or GPT) to control a browser like a human. It provides a set of tools for navigation, interaction, data extraction, and tab management, making it suitable for developers who want to automate browser tasks through AI assistants.
How to use Mcp Browser Agent?
Install the package globally via npm (npm install -g @yubao2000/mcp-browser-agent), then start the server with the command mcp-browse-agent. Configure it in supporting MCP clients (Claude Desktop, Cursor, or OpenAI Codex) by adding a new server entry pointing to the same command. A configuration file .mcp-browser.json can be used to set options like viewport, headless mode, and data directory.
Key features of Mcp Browser Agent
- 30 tools for navigation, clicking, filling forms, and scrolling
- Extract text, HTML, screenshots, and PDFs from pages
- Manage tabs: open, switch, close, and iframe contexts
- Execute custom JavaScript and retrieve console logs
- Cookie persistence for maintaining login sessions
- Optional visible browser window for debugging or manual intervention
Use cases of Mcp Browser Agent
- Automate web form filling and login workflows with AI
- Scrape structured data from multiple pages and export results
- Monitor website changes by capturing periodic screenshots
- Test web application behavior by simulating user actions
- Perform research tasks that require multi-step browsing (e.g., search, extract, navigate)
FAQ from Mcp Browser Agent
What are the system requirements?
Node.js version 18 or higher is required. The server runs headless by default but can display a browser window when configured.
How are cookies managed?
Cookies are automatically saved to ./.browser-data/cookies.json and restored on restart, allowing sessions to persist across agent runs.
Can I see what the AI is doing in the browser?
Yes. Set the environment variable SHOW_BROWSER=true or add "showBrowser": true to .mcp-browser.json to make the browser window visible for debugging, captcha handling, or demos.
Does the server support concurrent browser instances?
The free version supports one concurrent tab; the Pro version (upcoming, $9.9/month) supports up to 10 concurrent tabs along with batch automation and encrypted cookie backups.
Where does browser data (cookies, profiles) get stored?
By default, data is stored in the ./.browser-data directory. This location can be changed via the dataDir configuration option or BROWSER_DATA_DIR environment variable.
「AI とエージェント」の他のコンテンツ
mcp-hfspace MCP Server 🤗
evalstateMCP Server to Use HuggingFace spaces, easy configuration and Claude Desktop mode.
Sequential Thinking Multi-Agent System (MAS)
FradSerAn advanced sequential thinking process using a Multi-Agent System (MAS) built with the Agno framework and served via MCP.
Model Context Protocol Server for Home Assistant
tevonsbA MCP server for Home Assistant
Unreal Engine Generative AI Support Plugin
prajwalshettydevUnreal Engine plugin for LLM/GenAI models & MCP UE5 server. OpenAI GPT-5, Deepseek R1, Claude Opus/Sonnet, Gemini 3, Grok 4, Alibaba Qwen, Kimi, ElevenLabs TTS, Inworld, OpenRouter, Groq, GLM, Ollama, Local, Meshy, Tripo, Hunyuan3D, Rodin, fal, Dashscope, Seedream. NPC AI, agenti
Web Agent Protocol
OTA-Tech-AI🌐Web Agent Protocol (WAP) - Record and replay user interactions in the browser with MCP support
コメント