Which integrations are available for this server?

Allows downloading videos from Bilibili, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from Facebook, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from Instagram, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from TikTok, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from Twitch, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from Vimeo, including extracting video info, listing formats, and downloading at specified quality. Allows downloading videos from YouTube, including extracting video info, listing formats, and downloading at specified quality.

How do I use media-agent?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@media-agent Download this video to D:/downloads at 720p: https://www.youtube.com/watch?v=jNQXAC9IVRw" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

media-agent

by leohux

Overview Schema Related Servers Score Discussions

Python

Local

MEDIA-AGENT A MCP video-download agent for WorkBuddy, Cursor, and other hosts

media-agent 让 AI 智能体真正去下载视频：你在 WorkBuddy / Cursor / Claude Desktop 里丢一条链接，Agent 会先解析标题和时长，再把文件存到你指定的目录。

它是什么

一层专门给大模型调用的 MCP / CLI 接口。人不用记一长串下载参数；Agent 只看到三个固定工具，返回 JSON。

适合	不适合
WorkBuddy / Cursor 里说「把这个视频存到 D 盘」	网页版通义千问直接挂本地工具
命令行解析、下载单条链接	整份播放列表、整频道扒取
B站 / 抖音 / YouTube / 小红书等公开视频	会员、DRM、付费课破解

安装

Release 包

到 Releases 下载最新 wheel：

pip install media_agent-0.1.3-py3-none-any.whl
pip install mcp

从源码

git clone https://github.com/leohux/media-agent.git
cd media-agent
pip install -e ".[default,agent]"

自检：

python -m media_agent --help
python -m media_agent extract "https://www.youtube.com/watch?v=jNQXAC9IVRw"

ok 为 true 即解析成功。

依赖

Python ≥ 3.10
建议安装 ffmpeg（合并音视频 / 抽音频）。没装也能下，会自动改用单文件格式
给 Agent 用：pip install mcp

接到 AI 工具（MCP）

本项目是 MCP stdio 服务。谁能配自定义 MCP，谁就能用。完整示例见 mcp.example.json。

把 python 换成你机器上的解释器路径。

{
  "mcpServers": {
    "media-agent": {
      "command": "python",
      "args": ["-m", "media_agent"],
      "env": {
        "MEDIA_AGENT_OUTPUT_DIR": "D:/downloads",
        "MEDIA_AGENT_ALLOWED_ROOT": "D:/downloads",
        "MEDIA_AGENT_COOKIES_FROM_BROWSER": "chrome"
      }
    }
  }
}

WorkBuddy / Cursor / Claude

产品	配置位置
WorkBuddy	`~/.workbuddy/mcp.json` 或项目内 `.workbuddy/mcp.json`
Cursor	MCP 设置页，或 `.cursor/mcp.json`
Claude Desktop	`claude_desktop_config.json`

配好后直接说：

解析这个视频，确认标题后再下到 D:/downloads，清晰度 720p。 https://www.bilibili.com/video/BV...

千问

能：把千问放进 WorkBuddy / Qwen-Agent 当模型
不能：通义千问网页版、阿里云百炼远程 SSE（本项目是本地 stdio）

三个工具

Agent 只能调这三项，不会直接拼下载命令。

工具	作用	何时用
`extract_video_info`	只解析，不下文件	先确认标题、时长、UP 主
`list_formats`	列出清晰度	用户指定了 720p / 只下音频
`download_video`	下载一条视频	用户明确要保存文件

三个工具都接受 cookies_from_browser / cookiefile。抖音等站点通常需要先在浏览器打开过该页。

download_video 参数

参数	说明
`url`	必须是 `http://` 或 `https://` 的单条链接
`output_dir`	保存目录，可省略
`quality`	`best` / `1080p` / `720p` / `480p` / `360p` / `audio_only`
`audio_only`	`true` 时只下音频
`cookies_from_browser`	从浏览器读 cookie，如 `chrome` / `edge` / `chrome:Profile 1`
`cookiefile`	Netscape 格式 cookie 文件路径

解析成功时大致返回：

{
  "ok": true,
  "error": null,
  "id": "jNQXAC9IVRw",
  "title": "Me at the zoo",
  "duration": 19,
  "uploader": "jawed",
  "webpage_url": "https://www.youtube.com/watch?v=jNQXAC9IVRw",
  "ext": "mp4",
  "extractor": "Youtube"
}

失败则 ok 为 false，原因在 error 里。

命令行

python -m media_agent extract "https://www.bilibili.com/video/BVxxxx"
python -m media_agent formats "https://www.youtube.com/watch?v=..."
python -m media_agent download "https://www.youtube.com/watch?v=..." --output-dir D:/downloads --quality 720p
python -m media_agent download "https://..." --output-dir D:/downloads --audio-only
python -m media_agent extract --cookies-from-browser chrome "https://www.douyin.com/video/..."
python -m media_agent   # 启动 MCP，给 Agent 用

支持哪些网站

与上游内核相同，约一千多个提取器。常见：

国内：B站、抖音、小红书、微博、优酷、爱奇艺、腾讯视频、芒果 TV、西瓜视频、AcFun、央视网、网易云、QQ 音乐
国外：YouTube、TikTok、Twitter/X、Instagram、Facebook、Twitch、Vimeo

完整名单：supportedsites.md

网站改版后个别站点可能失效。快手没有独立提取器。会员内容经常下不了。

安全限制

限制	行为
协议	只接受 `http` / `https`
播放列表	拒绝，请传单条视频链接
目录	`MEDIA_AGENT_ALLOWED_ROOT` 之内才能写
覆盖	默认不覆盖已有文件

环境变量	作用
`MEDIA_AGENT_OUTPUT_DIR`	默认保存目录
`MEDIA_AGENT_ALLOWED_ROOT`	路径白名单根目录
`MEDIA_AGENT_COOKIES_FROM_BROWSER`	默认从哪个浏览器读 cookie（`chrome` / `edge` / `firefox` …）
`MEDIA_AGENT_COOKIEFILE`	Netscape cookie 文件路径

常见问题

WorkBuddy 配了但不调用？
确认 command 能执行 python -m media_agent --help，并且已安装 mcp。

解析成功，YouTube 下载 403？
站点风控常见。可换其他平台或直链，或加上 --cookies-from-browser chrome（需已在该浏览器登录）。

抖音提示 Fresh cookies are needed？
先在 Chrome / Edge 打开过该视频页，再传 --cookies-from-browser chrome 或 edge。浏览器开着时 Windows 上有时读不到 cookie，可先关掉浏览器再试。也可导出 Netscape cookie 文件后用 --cookies 路径。

提示要 ffmpeg？
没装也会下（单文件格式）。要高清画质+音轨合并，请安装 ffmpeg。

能下播放列表吗？
不能。打开列表里的单集再传链接。

和 yt-dlp 什么关系？
内核在 yt_dlp/，产品名是 media-agent。不要对这个仓库跑上游自动更新。

开发

media_agent/          对外入口：python -m media_agent
yt_dlp/agent/         三个工具 + MCP 服务
yt_dlp/               下载内核
test/test_agent_tools.py
mcp.example.json
docs/banner.svg
docs/hero.png
docs/architecture.png

python -m unittest test.test_agent_tools -v

Agent / MCP / 文档问题请在 本仓库 开 Issue。网站解析坏了请到 yt-dlp/yt-dlp。

致谢与许可

下载能力来自 yt-dlp/yt-dlp，与官方无隶属关系。详见 NOTICE。

许可证：Unlicense

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

No tool schema history has been recorded yet.

This server cannot be installed

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Maintenance

ActivityMaintained

ResponsivenessNo issues

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

mcp-hailuoOAuth
io.github.AceDataCloud
MCP server for Hailuo (MiniMax) AI video generation
AI & Machine Learning Image & Video Processing
21
mcp-wanOAuth
io.github.AceDataCloud
MCP server for Wan AI video generation
AI & Machine Learning Image & Video Processing
21
YouTube Transcript Downloader A
com.youtubetranscriptdownload
An MCP server that gives any LLM or agent clean YouTube transcripts on demand: a single video, a whole channel, or a playlist, plus AI cleanup of auto-generated captions. API-key auth, credit-based, same backend as the public v1 API. Get a free API key with 25 free credits at youtubetranscriptdownload.com/account.
Entertainment & Media Research & Data Web Scraping
42
clipkit A
dev.clipkit
MCP server for Clipkit — gives AI agents a video toolbox via the Clipkit schema.
Multimedia Processing Image & Video Processing
15215

Related MCP Servers

Video Fetch MCP
Multimedia Processing Web Scraping App Automation
pickstar-2002
-
license
A
quality
Not graded
maintenance
Enables downloading videos from 1000+ platforms including YouTube, Bilibili, TikTok, and Twitter using yt-dlp. Supports both MCP protocol and REST API modes with real-time progress tracking, multiple formats, and subtitle downloads.
Updated a year ago (2025-08-08 02:57 UTC)
5
14
1
-
video-toolkit-mcp
Multimedia Processing Web Scraping Audio Processing
JamesANZ
A
license
B
quality
D
maintenance
A Model Context Protocol (MCP) server that provides comprehensive video tools: transcript retrieval, video downloading, and automatic subtitle generation using AI speech-to-text. Works with YouTube, Bilibili, Vimeo, and any platform supported by yt-dlp.
Updated 4 months ago (2026-05-07 01:19 UTC)
11
MIT
MCP YouTube Downloader
Multimedia Processing Entertainment & Media
lucassnts963
A
license
Not graded
quality
D
maintenance
MCP server for downloading videos and audio from YouTube and hundreds of other sites using yt-dlp.
Updated 3 months ago (2026-06-04 01:13 UTC)
MIT
video-studio-mcp
AI & Machine Learning Image & Video Processing Social Media
Joshuayang228
F
license
Not graded
quality
B
maintenance
MCP server enabling AI agents to generate AI videos and images, analyze video content, and download videos from Douyin and Xiaohongshu.
Updated 2 days ago (2026-09-05 08:32 UTC)
-

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/leohux/media-agent'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

目录