YouTube 的模型上下文协议 (MCP) 服务器实现，使 AI 语言模型能够通过标准化界面与 YouTube 内容进行交互。
获取视频详细信息（标题、描述、时长等）
列出频道视频
获取视频统计数据（观看次数、点赞次数、评论次数）
在 YouTube 上搜索视频
检索视频记录
支持多种语言
获取带时间戳的字幕
在成绩单内搜索
获取频道详情
列出频道播放列表
获取频道统计信息
在频道内容内搜索
列出播放列表项目
获取播放列表详细信息
在播放列表中搜索
获取播放列表视频脚本
Related MCP server: mcp-youtube-transcript
要通过Smithery自动为 Claude Desktop 安装 YouTube MCP 服务器：
npx -y @smithery/cli install @ZubeidHendricks/youtube --client claudenpm install zubeid-youtube-mcp-server设置以下环境变量：
YOUTUBE_API_KEY ：您的 YouTube 数据 API 密钥
YOUTUBE_TRANSCRIPT_LANG ：转录的默认语言（可选，默认为“en”）
将其添加到您的 MCP 客户端配置（例如 Claude Desktop）：
{
"mcpServers": {
"youtube": {
"command": "npx",
"args": ["-y", "zubeid-youtube-mcp-server"],
"env": {
"YOUTUBE_API_KEY": "<YOUR_API_KEY>"
}
}
}
}对于一键安装，请单击以下安装按钮之一：
如果您希望手动安装，请先勾选本节顶部的安装按钮。否则，请按照以下步骤操作：
将以下 JSON 块添加到 VS Code 中的“用户设置 (JSON)”文件中。您可以按下Ctrl + Shift + P并输入Preferences: Open User Settings (JSON)来执行此操作。
{
"mcp": {
"inputs": [
{
"type": "promptString",
"id": "apiKey",
"description": "YouTube API Key",
"password": true
}
],
"servers": {
"youtube": {
"command": "npx",
"args": ["-y", "zubeid-youtube-mcp-server"],
"env": {
"YOUTUBE_API_KEY": "${input:apiKey}"
}
}
}
}
}或者，您可以将其添加到工作区中名为.vscode/mcp.json的文件中：
{
"inputs": [
{
"type": "promptString",
"id": "apiKey",
"description": "YouTube API Key",
"password": true
}
],
"servers": {
"youtube": {
"command": "npx",
"args": ["-y", "zubeid-youtube-mcp-server"],
"env": {
"YOUTUBE_API_KEY": "${input:apiKey}"
}
}
}
}前往 Google Cloud Console
创建新项目或选择现有项目
启用 YouTube 数据 API v3
创建 API 凭证（API 密钥）
复制 API 密钥进行配置
// Get video details
const video = await youtube.videos.getVideo({
videoId: "video-id"
});
// Get video transcript
const transcript = await youtube.transcripts.getTranscript({
videoId: "video-id",
language: "en"
});
// Search videos
const searchResults = await youtube.videos.searchVideos({
query: "search term",
maxResults: 10
});// Get channel details
const channel = await youtube.channels.getChannel({
channelId: "channel-id"
});
// List channel videos
const videos = await youtube.channels.listVideos({
channelId: "channel-id",
maxResults: 50
});// Get playlist items
const playlistItems = await youtube.playlists.getPlaylistItems({
playlistId: "playlist-id",
maxResults: 50
});
// Get playlist details
const playlist = await youtube.playlists.getPlaylist({
playlistId: "playlist-id"
});# Install dependencies
npm install
# Run tests
npm test
# Build
npm run build
# Lint
npm run lint有关对此存储库做出贡献的信息，请参阅 CONTRIBUTING.md。
该项目根据 MIT 许可证获得许可 - 有关详细信息，请参阅 LICENSE 文件。
Search for videos on YouTube
| Name | Required | Description | Default |
|---|---|---|---|
| order | No | Sort order for results, such as relevance or date | |
| query | Yes | Search query | |
| sortBy | No | Optional ranking mode such as relevance, indie_priority, subscribers_asc, subscribers_desc, or recent_activity | |
| channelId | No | Restrict results to a specific channel ID | |
| maxResults | No | Maximum number of results to return | |
| creatorOnly | No | Only include channels heuristically classified as creators | |
| publishedAfter | No | Only include videos published after this ISO 8601 date | |
| uniqueChannels | No | Return only one matched video per unique channel | |
| publishedBefore | No | Only include videos published before this ISO 8601 date | |
| channelMaxSubscribers | No | Maximum subscriber count for the matched video channel | |
| channelMinSubscribers | No | Minimum subscriber count for the matched video channel | |
| channelLastUploadAfter | No | Only include videos whose channel latest upload is after this ISO 8601 date | |
| channelLastUploadBefore | No | Only include videos whose channel latest upload is before this ISO 8601 date |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
There are no annotations, so the description must disclose behavioral traits. It only says 'Search for videos,' omitting any details about rate limits, pagination, return format, or side effects. This is insufficient for an agent to anticipate behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence, which is too brief for a tool with 13 parameters. While concise, it is under-specified and does not earn its place; it should include more context.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (13 parameters, no output schema), the description lacks completeness. It does not explain return values, default behaviors, or interaction effects between parameters, leaving the agent underinformed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, so the schema itself documents each parameter. The tool description adds no additional meaning beyond the schema, making the baseline score of 3 appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states a specific verb ('Search') and resource ('videos on YouTube'), making the tool's purpose immediately understandable. However, it lacks any differentiation from potential sibling tools (none provided) and is somewhat generic.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool, when not to use it, or any alternatives. With no sibling tools mentioned, the description could at least imply typical search scenarios, but it does not.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.2channels_getChannelchannels_listVideosplaylists_getPlaylistplaylists_getPlaylistItemstranscripts_getTranscriptvideos_getVideovideos_searchVideos11 fields changedInput schema / properties / channelIdAdded value: +{
+ "description": "Restrict results to a specific channel ID",
+ "type": "string"
+}Input schema / properties / channelLastUploadAfterAdded value: +{
+ "description": "Only include videos whose channel latest upload is after this ISO 8601 date",
+ "type": "string"
+}Input schema / properties / channelLastUploadBeforeAdded value: +{
+ "description": "Only include videos whose channel latest upload is before this ISO 8601 date",
+ "type": "string"
+}Input schema / properties / channelMaxSubscribersAdded value: +{
+ "description": "Maximum subscriber count for the matched video channel",
+ "type": "number"
+}Input schema / properties / channelMinSubscribersAdded value: +{
+ "description": "Minimum subscriber count for the matched video channel",
+ "type": "number"
+}Input schema / properties / creatorOnlyAdded value: +{
+ "description": "Only include channels heuristically classified as creators",
+ "type": "boolean"
+}Input schema / properties / orderAdded value: +{
+ "description": "Sort order for results, such as relevance or date",
+ "type": "string"
+}Input schema / properties / publishedAfterAdded value: +{
+ "description": "Only include videos published after this ISO 8601 date",
+ "type": "string"
+}Input schema / properties / publishedBeforeAdded value: +{
+ "description": "Only include videos published before this ISO 8601 date",
+ "type": "string"
+}Input schema / properties / sortByAdded value: +{
+ "description": "Optional ranking mode such as relevance, indie_priority, subscribers_asc, subscribers_desc, or recent_activity",
+ "type": "string"
+}Input schema / properties / uniqueChannelsAdded value: +{
+ "description": "Return only one matched video per unique channel",
+ "type": "boolean"
+}v1.0.0channels_getChannelchannels_listVideosplaylists_getPlaylistplaylists_getPlaylistItemstranscripts_getTranscriptvideos_getVideovideos_searchVideosScored across 1 tool
With only one tool, there is no possibility of confusion or overlap between tools, so disambiguation is perfect.
The single tool name 'videos_searchVideos' mixes camelCase and an awkward verb_noun structure ('videos' as prefix then 'searchVideos'), showing internal inconsistency and lack of a clear pattern.
One tool is far too few for a YouTube MCP server, which typically requires multiple operations (search, video details, channel info, etc.), making the tool count extremely inadequate.
The tool only supports searching for videos, missing critical operations like retrieving video details, listing playlists, or managing subscriptions, leaving severe gaps in covering the YouTube domain.
YouTube transcripts, search, channel/playlist listings and upload tracking for AI agents.
YouTube transcripts, search, and channel and playlist listings for any AI app. Hosted.
YouTube transcripts, search, channel browsing, and playlists for AI agents via MCP.
YouTube for AI agents: transcripts, comments, video stats, channel monitor, Shorts captions.
