🎉 MediaGo v3.5.0

This release brings support for more sites, a brand-new browser extension, and deep integration with AI assistants. Overall, it runs faster and uses fewer resources.

🌐 Browser Extension (Chrome / Edge)

See a video you like while browsing? Send it to MediaGo with one click.

• Automatically detects downloadable videos on web pages — the toolbar icon shows the count detected on the current page
• Works with most major video sites, including Bilibili, YouTube, and more
• Bilingual interface (English / Chinese) that follows your browser language by default
• No need to install from the store — after installing MediaGo, find the extension files under Settings → More Settings → Browser Extension Directory

🎬 YouTube and 1000+ Sites Supported

Added support for YouTube, Twitter/X, Instagram, Reddit, and over a thousand other video sites. Previously supported formats are now more stable too, especially for overseas sites.

🦞 Download with a Single Sentence via AI Assistants (OpenClaw Skill)

If you use AI coding tools like Claude Code or Cursor, you can now just tell your AI:

"Download this video for me: https://..."

The AI will automatically call MediaGo to handle the download. To install:

npx clawhub@latest install mediago

Both English and Chinese commands are recognized.

🔌 Let Other Tools Integrate with MediaGo

A full set of APIs is now open, letting scripts, automation tools, and even other apps call MediaGo — you're no longer confined to the desktop window. The browser extension itself communicates with the desktop app through these APIs, and power users can build them into their own workflows.

🎞️ Convert Formats Right After Downloading

Once a video finishes downloading, you can convert its format and choose quality settings directly inside MediaGo — no need to open another tool.

🐳 Easier Docker Deployment

Deploy with a single command. Works on both Intel and ARM chips:

docker run -d --name mediago -p 8899:8899 \
  -v /path/to/mediago:/app/mediago \
  ghcr.io/caorushizi/mediago:3.5.0

If you're using the desktop version, phones and tablets on the same home network can also open MediaGo's web interface — no extra configuration needed.

⚡ Performance and Experience

• Faster startup and lower memory footprint
• Built-in video player — no need to launch a separate app
• Some long-standing issues are fixed: download logs can now be copied, the download task list auto-refreshes when tasks are imported externally, and special characters in filenames are handled automatically — no more "can't find the file after it finishes downloading"

Notes for macOS Users

[Intel chips] Install the x64 version from the release page. After installation, you'll need to allow apps from unidentified developers in your Mac's security settings.

[Apple Silicon] Install the arm64 version from the release page. After installation, run the following command in Terminal:

sudo xattr -dr com.apple.quarantine /Applications/mediago-community.app

🎉 MediaGo v3.5.0

新版本带来更多站点支持、全新浏览器扩展、以及和 AI 助手的深度整合。整体运行更快、资源占用更少。

🌐 浏览器扩展（Chrome / Edge）

浏览网页时看到感兴趣的视频，一键发到 MediaGo 下载。

自动识别网页上的可下载视频，工具栏图标会显示当前页检测到的数量
支持大多数主流视频站点 —— 包括 B站、YouTube 等
中英双语界面，默认跟随浏览器语言
不用去商店下载，装完 MediaGo 后在设置 → 更多设置 → 浏览器扩展目录就能找到安装文件

🎬 支持 YouTube 和 1000+ 站点

新增对 YouTube、Twitter/X、Instagram、Reddit 等一千多个视频站点的支持。之前只能下的格式也更稳定了，特别是一些海外网站。

🦞 AI 助手一句话下载（OpenClaw Skill）

如果你在用 Claude Code、Cursor 等 AI 编程工具，现在可以直接跟 AI 说：

"帮我下载这个视频：https://..."

AI 会自动调用 MediaGo 帮你下载。安装方式：

npx clawhub@latest install mediago

中文英文指令都能识别。

🔌 让其他工具和 MediaGo 联动

开放了一整套接口，让脚本、自动化工具、甚至其他 App 都能调用 MediaGo —— 你不再被限制在桌面窗口里。浏览器扩展就是通过这套接口和桌面端对话的，喜欢折腾的用户也可以基于它接入自己的工作流。

🎞️ 下载完直接转格式

视频下载好之后，在 MediaGo 里直接就能转格式、选画质，不用再打开别的软件。

🐳 Docker 部署更简单

一条命令就能部署，支持 Intel 和 ARM 两种芯片：

docker run -d --name mediago -p 8899:8899 \
  -v /path/to/mediago:/app/mediago \
  ghcr.io/caorushizi/mediago:3.5.0

如果你用的是桌面版，同一个家庭网络里的手机、平板也能打开 MediaGo 的 Web 界面，不用再额外配置。

⚡ 速度与体验

启动更快、占用更少内存
视频播放器内置了，不再需要单独启动
一些老顽疾修掉了：下载日志现在可以复制、下载任务列表在外部导入时会自动刷新、文件名里的特殊字符会自动处理，不会再出现"下完找不到文件"的问题

MacOS 使用

【Intel 芯片】需要安装 release 中的 x64 版本。安装完成后需要在 Mac 的安全设置中打开来自身份不明开发者的 Mac App
【Apple 芯片】需要安装 release 中的 arm64 芯片。安装完成后需要在控制台执行 sudo xattr -dr com.apple.quarantine /Applications/mediago-community.app 命令