YouTube Toolbox
py-mcp-youtube-工具箱
MCP 服务器为 AI 助手提供与 YouTube 交互的强大工具,包括视频搜索、文字记录提取、评论检索等。
概述
py-mcp-youtube-toolbox 提供以下与 YouTube 相关的功能:
使用高级过滤选项搜索 YouTube 视频
获取有关视频和频道的详细信息
使用排序选项检索视频评论
提取多种语言的视频记录和字幕
查找给定视频的相关视频
按地区获取热门视频
根据文字记录生成视频内容摘要
具有过滤、搜索和多视频功能的高级转录分析
Related MCP server: YouTube MCP Server
目录
先决条件
Python :安装 Python 3.12 或更高版本
YouTube API 密钥:
创建新项目或选择现有项目
启用 YouTube 数据 API v3:
前往“API 和服务”>“库”
搜索并启用“YouTube 数据 API v3”
创建凭证:
前往“API 和服务”>“凭证”
点击“创建凭证”>“API 密钥”
记下您的 API 密钥
安装
Git 克隆
git clone https://github.com/jikime/py-mcp-youtube-toolbox.git
cd py-mcp-youtube-toolbox配置
安装 UV 包管理器:
curl -LsSf https://astral.sh/uv/install.sh | sh创建并激活虚拟环境:
uv venv -p 3.12
source .venv/bin/activate # On MacOS/Linux
# or
.venv\Scripts\activate # On Windows安装依赖项:
uv pip install -r requirements.txt环境变量:
cp env.example .env
vi .env
# Update with your YouTube API key
YOUTUBE_API_KEY=your_youtube_api_key使用 Docker
构建 Docker 镜像:
docker build -t py-mcp-youtube-toolbox .运行容器:
docker run -e YOUTUBE_API_KEY=your_youtube_api_key py-mcp-youtube-toolbox使用本地
运行服务器:
mcp run server.py运行 MCP 检查器:
mcp dev server.py配置 MCP 设置
将服务器配置添加到您的 MCP 设置文件:
克劳德桌面应用程序
要通过Smithery自动安装:
npx -y @smithery/cli install @jikime/py-mcp-youtube-toolbox --client claude要手动安装,请打开
~/Library/Application Support/Claude/claude_desktop_config.json
将其添加到mcpServers对象:
{
"mcpServers": {
"YouTube Toolbox": {
"command": "/path/to/bin/uv",
"args": [
"--directory",
"/path/to/py-mcp-youtube-toolbox",
"run",
"server.py"
],
"env": {
"YOUTUBE_API_KEY": "your_youtube_api_key"
}
}
}
}游标 IDE
打开~/.cursor/mcp.json
将其添加到mcpServers对象:
{
"mcpServers": {
"YouTube Toolbox": {
"command": "/path/to/bin/uv",
"args": [
"--directory",
"/path/to/py-mcp-youtube-toolbox",
"run",
"server.py"
],
"env": {
"YOUTUBE_API_KEY": "your_youtube_api_key"
}
}
}
}对于 Docker
{
"mcpServers": {
"YouTube Toolbox": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "YOUTUBE_API_KEY=your_youtube_api_key",
"py-mcp-youtube-toolbox"
]
}
}
}工具文档
视频工具
search_videos:使用高级过滤选项(频道、时长、地区等)搜索 YouTube 视频get_video_details:获取有关特定 YouTube 视频的详细信息(标题、频道、观看次数、喜欢次数等)get_video_comments:从 YouTube 视频中检索带有排序选项的评论get_related_videos:查找与特定 YouTube 视频相关的视频get_trending_videos:按地区获取 YouTube 上的热门视频
渠道工具
get_channel_details:获取有关 YouTube 频道的详细信息(名称、订阅者、观看次数等)
成绩单工具
get_video_transcript:从指定语言的 YouTube 视频中提取文字记录/字幕get_video_enhanced_transcript:具有过滤、搜索和多视频功能的高级转录提取
提示工具
transcript_summary:根据可自定义选项的成绩单生成 YouTube 视频内容摘要
资源工具
youtube://available-youtube-tools:获取所有可用的 YouTube 工具列表youtube://video/{video_id}:获取特定视频的详细信息youtube://channel/{channel_id}:获取有关特定频道的信息youtube://transcript/{video_id}?language={language}:获取特定视频的成绩单
发展
对于本地测试,您可以使用附带的客户端脚本:
# Example: Search videos
uv run client.py search_videos query="MCP" max_results=5
# Example: Get video details
uv run client.py get_video_details video_id=zRgAEIoZEVQ
# Example: Get channel details
uv run client.py get_channel_details channel_id=UCRpOIr-NJpK9S483ge20Pgw
# Example: Get video comments
uv run client.py get_video_comments video_id=zRgAEIoZEVQ max_results=10 order=time
# Example: Get video transcript
uv run client.py get_video_transcript video_id=zRgAEIoZEVQ language=ko
# Example: Get related videos
uv run client.py get_related_videos video_id=zRgAEIoZEVQ max_results=5
# Example: Get trending videos
uv run client.py get_trending_videos region_code=ko max_results=10
# Example: Advanced transcript extraction
uv run client.py get_video_enhanced_transcript video_ids=zRgAEIoZEVQ language=ko format=timestamped include_metadata=true start_time=100 end_time=200 query=에이전트 case_sensitive=true segment_method=equal segment_count=2
# Example: 执照
MIT 许可证
Available Tools
8 toolsget_channel_detailsB
Get detailed information about a YouTube channel
Inputs: channel_id (required). Returns: result
get_trending_videosB
Get trending videos on YouTube by region
Inputs: region_code, max_results. Returns: result
get_video_commentsC
Get comments for a YouTube video
Inputs: video_id (required), max_results, order, include_replies, +1 more. Returns: result
get_video_detailsB
Get detailed information about a YouTube video
Inputs: video_id (required). Returns: result
get_video_enhanced_transcriptC
Advanced transcript extraction tool with filtering, search, and multi-video capabilities. Provides rich transcript data for detailed analysis and processing. Features: 1) Extract transcripts from multiple videos; 2) Filter by time ranges; 3) Search within transcripts; 4) Segment transcripts; 5) Format output in different ways; 6) Include video metadata.
Inputs: video_ids (required), language, start_time, end_time, +6 more. Returns: result
get_video_transcriptB
Get transcript/captions for a YouTube video
Inputs: video_id (required), language. Returns: result
search_videosC
Search for YouTube videos with advanced filtering options
Inputs: query (required), max_results, channel_id, order, +6 more. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 8 tools
Most tools have distinct purposes, but there is notable overlap between get_video_transcript and get_video_enhanced_transcript, which could cause confusion as both handle transcripts with the enhanced version being a superset. Other tools like get_video_details and get_channel_details are clearly differentiated, but the transcript duplication weakens clarity.
All tool names follow a consistent snake_case pattern with a verb-noun structure (e.g., get_channel_details, search_videos). The naming is predictable and uniform across all eight tools, making it easy for agents to understand and use them without confusion.
With 8 tools, the count is well within the typical 3-15 range for a focused domain like YouTube data retrieval. It feels slightly thin for comprehensive coverage but reasonable for core functionalities such as fetching videos, channels, comments, and transcripts.
The toolset covers key read operations like getting details, searching, and fetching comments/transcripts, but lacks obvious write or management capabilities (e.g., upload, update, delete) that might be expected in a full YouTube toolbox. This creates notable gaps for agents needing to perform more than retrieval tasks.
Maintenance
Related MCP Connectors
An MCP server that integrates with Discord to provide AI-powered features.
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.
Driflyte MCP server which lets AI assistants query topic-specific knowledge from web and GitHub.
Fetch YouTube transcripts, monitor channels and track keyword matches from your AI assistant.
Related MCP Servers
- AlicenseAqualityCmaintenanceAn MCP server for intelligent YouTube video analysis that provides token-optimized summaries, sentiment analysis, and entity extraction from transcripts. It enables AI assistants to perform video reporting, channel monitoring, and comprehensive YouTube searches through structured data tools.1053Apache 2.0
- AlicenseAqualityBmaintenanceMCP server that provides YouTube video data to AI agents, supporting search, metadata, comments, and transcripts without an API key.681 npmISC
- AlicenseAqualityDmaintenanceMCP server that lets AI agents search YouTube and fetch transcripts.23MIT
- FlicenseNot gradedqualityDmaintenanceMCP server for YouTube that provides tools to fetch video metadata and transcripts, enabling natural language queries about YouTube videos.2-