
概览
主要功能
- 支持浏览器互动的 MCP 服务器
- 支持 M2 和 M3 浏览器版本
- 兼容 Chromium-Based 浏览器、Firefox 和 Safari
- 支持在 Fronteir AI 上的托管部署
- 支持 MCP 客户端
价格
- 模型
- Free
- 分类
- 服务配缩器
- 评分
- 暂无评价
使用场景
AI 助手集成
使用 MCP 浏览器套件将 AI 助手集成到本地浏览器中,开启交互和自动化。
浏览器插件开发
使用 MCP 浏览器套件开发和加载浏览器插件,支持各种浏览器版本和类型。
优点 & 缺点
优点
- 让 AI 助手能够与本地浏览器互动
- 支持多种浏览器版本(M2 和 M3)
- 支持在 Fronteir AI 上的托管部署
缺点
- 当前缺少对 incoming 请求的身份验证
- 需要用户保证端口安全性,确保在本机网络上只允许特定端口
评测
登录以留下评测。
暂无评测。来当第一个吧!
问答
Can I host the MCP Browser Kit on a cloud service?
Yes, the kit is available for hosted deployment on Frontier AI, allowing you to run the MCP Server in the cloud while still connecting to local browsers via the extension.
Asked by Celia Ramirez · Apr 5, 2026
Is there any authentication for incoming requests to the MCP Server?
No, the kit currently does not implement authentication for incoming requests. Users should restrict the server’s ports to their local device network to prevent unauthorized access.
Asked by Dmitri Volkov · Mar 24, 2026
What is the difference between the M2 and M3 extension builds?
M2 provides a richer set of capabilities for the AI assistant to interact with your browser, while M3 offers a slimmer feature set. Choose M2 if you need full functionality; M3 may be sufficient for simpler tasks.
Asked by Elif Yildiz · Mar 21, 2026
Which browsers can I use with the MCP Browser Kit and how do I install the extension?
The kit works with Chrome, Brave, Edge, Firefox, and Safari. You add the kit to your MCP Client, download the matching browser extension (M2 for more features, M3 for basic), follow the specific installation steps for your browser, enable the extension, and refresh any open tabs.
Asked by Salome Beridze · Feb 19, 2026
提问
服务配缩器 的替代品

开源MCP服务器,利用_playwright让LLM控制真实的浏览器以及可访问性快照

Python agent 框架由 Pydantic 团队开发,用于构建类型安全的 GenAI 应用。

自适应的记忆层,帮助AI智能体从上下文中学习

AI 邮箱助手,帮助您更快地达到 inbox zero。

开源式24小时设备屏幕和音频录制,适合构建依赖实况人工智能(AI)应用

用于构建和编排带有工具、内存和多智能体工作流的AI智能体的TypeScript库。

把无银行的 onchain API 带入 MCP

带少分享系兿司上字詰級下合存号来皅




