MCP YouTube Transcript
An MCP server retrieving transcripts of YouTube videos
作者提供的说明(英文原文)
Structured public data from YouTube, TikTok, Instagram, Reddit, X, Facebook, Google Maps and more.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
适用于网页版、桌面版和手机版 Claude:
https://mcp.scrapingdaddy.com在终端中运行这条命令
claude mcp add --transport http scraping-daddy https://mcp.scrapingdaddy.com粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"scraping-daddy": {
"url": "https://mcp.scrapingdaddy.com"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"scraping-daddy": {
"type": "http",
"url": "https://mcp.scrapingdaddy.com"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
ChatGPT 只能连接互联网上的服务器:
https://mcp.scrapingdaddy.com菜单名称可能因套餐和版本不同而略有差异。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
工具名称和描述与服务器在 2026年11月6日 报告的完全一致。这些内容由作者撰写,不是我们写的;我们以纯文本形式显示。
list_platformsList every platform Scraping Daddy ("SD") can scrape, with a short description. Start here when the user asks SD for live data and the endpoint is not known yet. Discovery only — this does not consume API credits.
list_endpointsFind AM-CORE endpoints, optionally filtered by platform and/or a search term.
get_endpoint_schemaReturn the full request and response schema for one AM-CORE endpoint.
api_requestCall any Scraping Daddy ("SD") endpoint and return its response unchanged.
tiktok_profileGet a TikTok creator's profile: bio, follower/following counts, and stats.
tiktok_videoGet TikTok video or photo-post data by URL.
instagram_profileGet an Instagram profile: bio, follower counts, and account metadata.
instagram_mediaGet a single Instagram post or reel by its shortcode.
facebook_pageGet a Facebook page's profile data.
reddit_searchSearch Reddit posts.
youtube_videoGet YouTube video or Short details by URL.
x_profileGet a public X profile by handle or x.com profile URL.
x_tweetsGet raw public X User and Tweet objects for a handle or profile URL.
x_tweetGet one raw public X Tweet object by ID or status URL.
google_maps_searchSearch places or select one by place_id/data. Offsets advance by 20 (recommended maximum 100); later pages require ll. Only AM-CORE's supported domains/data filters are accepted.
google_maps_placeGet business details using at least one Google feature ID, place ID, or decimal CID. Multiple identifiers must agree.
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过域名 scrapingdaddy.com 验证(注册中心会核实域名所有权)
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
在 367 毫秒内响应(HTTP 200)
无需登录
已读取 16 个工具描述,未发现针对 AI 的隐藏指令
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在作者的云端运行。通过互联网连接,无需安装,但你发送的内容会经过作者的服务器。
GitHub 仓库最后更新于–,最新版本(1.0.0)于5天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。
同一分类中有:MCP YouTube Transcript, Google Analytics MCP, BibiGPT YouTube Video Summarizer
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
An MCP server retrieving transcripts of YouTube videos
GA4 MCP server for AI agents: schema discovery, server-side aggregation, analysis-ready defaults.
AI video summarizer: summarize and transcribe YouTube, Bilibili, TikTok and podcast videos.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Scraping Daddy MCP Server
A hosted social media scraping MCP server. Ask Claude, ChatGPT, Cursor or VS Code
for live public data from TikTok, Instagram, YouTube, Reddit, X and 20+ platforms.
Website · Docs · All 180+ endpoints
Server URL: https://mcp.scrapingdaddy.com
The server teaches your assistant that "SD" means Scraping Daddy, so you can just ask:
The assistant picks the right endpoint, calls it with your account's credits and answers from the live data.
Claude
https://mcp.scrapingdaddy.com.ChatGPT
https://mcp.scrapingdaddy.com under Server URL and keep Authentication on OAuth.Claude Code
claude mcp add --transport http scrapingdaddy-mcp https://mcp.scrapingdaddy.com
Then type /mcp, pick scrapingdaddy-mcp and choose Authenticate.
Cursor
Open Cursor Settings → Tools & MCP, click New MCP Server and add:
{
"mcpServers": { "scrapingdaddy-mcp": { "url": "https://mcp.scrapingdaddy.com" } } }
Save, then click Connect next to scrapingdaddy-mcp.
VS Code
https://mcp.scrapingdaddy.com and name it scrapingdaddy-mcp.Other MCP clients
Add a remote MCP server with the URL https://mcp.scrapingdaddy.com. If asked, choose Streamable HTTP and OAuth.
TikTok and TikTok Shop, Instagram, YouTube, X (Twitter), Reddit, Facebook, LinkedIn, Google Maps, Pinterest, Snapchat, Spotify, Twitch, Bluesky, GitHub, SoundCloud, Truth Social, Tumblr, Kick, Rumble and Linktree. Public data only.
| Tool | What it does | Credits |
|---|---|---|
list_platforms | Lists every platform SD can read | Free |
list_endpoints | Finds endpoints by platform or keyword | Free |
get_endpoint_schema | Shows an endpoint's parameters and response shape | Free |
api_request | Calls any of the 195 public endpoints and returns the response unchanged | Per call |
| Shortcuts | TikTok profile and video, Instagram profile and post, Facebook page, Reddit search, YouTube video, X profile, posts and single post, Google Maps search and place | Per call |
Finding platforms and endpoints is free. Scraping calls use the credits and limits of your Scraping Daddy account, the same as API requests. One question can lead to several calls.
See or disconnect connected apps at any time in your dashboard under Connected apps: https://scrapingdaddy.com/dashboard/connected-apps.
This repository documents the hosted server and holds its registry metadata (server.json). The server itself runs at https://mcp.scrapingdaddy.com; there is no code to install.
Prefer plain HTTP? The same data is available as a REST API: see the API reference.
Email support@scrapingdaddy.com or open an issue in this repository.