Skip to main content
Glama
enesjakozh
nattyraz

YouTube MCP Server

by nattyraz
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Servidor MCP de YouTube

Un servidor de Protocolo de Contexto de Modelo (MCP) para interactuar con vídeos de YouTube. Este servidor proporciona herramientas para extraer metadatos y subtítulos de vídeo, y convertirlos a formato Markdown con diversas plantillas.

Características

  • Metadatos de video : obtenga información completa del video

  • Extracción de subtítulos : compatibilidad con subtítulos generados automáticamente y manualmente

  • Varios idiomas : soporte integrado para inglés y francés

  • Sistema de plantillas : tres plantillas de rebajas integradas:

    • Básico: Formato de transcripción simple

    • Detallado: metadatos completos con marcas de tiempo

    • Búsqueda: Resultados resaltados con contexto

  • Función de búsqueda : Buscar dentro de los subtítulos de los videos

  • Autenticación flexible : admite tanto la clave API como la autenticación OAuth2

Related MCP server: YouTube Video Summarizer MCP Server

Prerrequisitos

  • Node.js (v16 o superior)

  • npm o hilo

  • Una clave de API de datos de YouTube y/o credenciales OAuth2

Instalación

  1. Clonar el repositorio:

git clone [repository-url]
cd youtube-mcp

  1. Instalar dependencias:

npm install

  1. Construir el proyecto:

npm run build

Configuración

Crea un archivo .env en el directorio raíz con tus credenciales de YouTube:

YOUTUBE_API_KEY=your_api_key
YOUTUBE_CLIENT_ID=your_client_id
YOUTUBE_CLIENT_SECRET=your_client_secret
YOUTUBE_REFRESH_TOKEN=your_refresh_token  # Optional, for OAuth2

Configuración de MCP

Agregue el servidor a su archivo de configuración de MCP (generalmente en ~/.config/Code/User/globalStorage/rooveterinaryinc.roo-cline/settings/cline_mcp_settings.json ):

{
  "mcpServers": {
    "youtube": {
      "command": "node",
      "args": ["path/to/youtube-mcp/build/index.js"],
      "env": {
        "YOUTUBE_API_KEY": "your_api_key",
        "YOUTUBE_CLIENT_ID": "your_client_id",
        "YOUTUBE_CLIENT_SECRET": "your_client_secret"
      },
      "disabled": false,
      "alwaysAllow": []
    }
  }
}

Uso

El servidor proporciona las siguientes herramientas:

1. Obtener información del video

use_mcp_tool youtube get_video_info {
  "url": "https://www.youtube.com/watch?v=VIDEO_ID"
}

2. Obtener subtítulos

use_mcp_tool youtube get_captions {
  "url": "https://www.youtube.com/watch?v=VIDEO_ID",
  "language": "en"  // Optional, defaults to "en"
}

3. Convertir a Markdown

use_mcp_tool youtube convert_to_markdown {
  "url": "https://www.youtube.com/watch?v=VIDEO_ID",
  "template_name": "detailed",  // Optional, "basic", "detailed", or "search"
  "language": "en",            // Optional
  "options": {                 // Optional
    "include_chapters": true,
    "search_term": "keyword"   // Only for search template
  }
}

4. Plantillas de lista

use_mcp_tool youtube list_templates

Dependencias

{
  "dependencies": {
    "@modelcontextprotocol/sdk": "latest",
    "googleapis": "^146.0.0",
    "google-auth-library": "^9.0.0",
    "youtube-captions-scraper": "^2.0.0",
    "express": "^4.18.2",
    "open": "^9.1.0"
  },
  "devDependencies": {
    "@types/node": "^20.0.0",
    "typescript": "^5.0.0",
    "tsx": "^4.0.0"
  }
}

Configuración de OAuth2

Para la autenticación OAuth2 (necesaria para el acceso a video privado):

  1. Crear un proyecto en Google Cloud Console

  2. Habilitar la API de datos de YouTube v3

  3. Crear credenciales OAuth2 (tipo de aplicación web)

  4. Ejecute el script de autenticación:

node src/get-api-key.js

  1. Siga las instrucciones del navegador para autorizar la aplicación.

  2. Copia el token de actualización a tu configuración

Personalización de plantillas

Puedes agregar plantillas personalizadas modificando la matriz DEFAULT_TEMPLATES en src/index.ts . Las plantillas siguen esta estructura:

interface MarkdownTemplate {
  name: string;
  description: string;
  format: {
    header?: string;
    chapter_format?: string;
    caption_block: string;
    timestamp_format?: string;
    search_result_format?: string;
  }
}

Licencia

Instituto Tecnológico de Massachusetts (MIT)

Contribuyendo

  1. Bifurcar el repositorio

  2. Crea tu rama de funciones ( git checkout -b feature/amazing-feature )

  3. Confirme sus cambios ( git commit -m 'Add some amazing feature' )

  4. Empujar a la rama ( git push origin feature/amazing-feature )

  5. Abrir una solicitud de extracción

Available Tools

1 tool
get_transcriptC

YouTubeの字幕をURLまたは動画IDから取得します

ParametersJSON Schema
NameRequiredDescriptionDefault
urlYesYouTube動画のURLまたはID
langYes字幕の言語コード（例: 'ja', 'en'）en

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full burden for behavioral disclosure. It mentions retrieving transcripts but doesn't describe what happens if no transcript exists, rate limits, authentication requirements, error conditions, or the format/scope of returned data. This leaves significant behavioral gaps for a tool that interacts with external APIs.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, efficient sentence in Japanese that clearly communicates the core functionality without any wasted words. It's appropriately sized and front-loaded with the essential information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with no annotations, no output schema, and interaction with an external service (YouTube), the description is insufficient. It doesn't explain what format the transcript returns, error handling, availability constraints, or any behavioral aspects needed for reliable use by an AI agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already fully documents both parameters (url and lang with their descriptions and default). The description adds no additional parameter semantics beyond what's in the schema, maintaining the baseline score for high schema coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the action ('取得します' - get/retrieve) and resource ('YouTubeの字幕' - YouTube subtitles/transcripts) with the input source specified ('URLまたは動画IDから' - from URL or video ID). It's specific about what the tool does, though it doesn't need to distinguish from siblings since there are none.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

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 versus alternatives, prerequisites, or limitations. It simply states what the tool does without context about appropriate use cases or constraints.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool update
    • First observedget_transcript
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

B3/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool has a clear and distinct purpose focused on retrieving transcripts from YouTube.

Naming Consistency5/5

A single tool inherently has perfect naming consistency as there are no other tools to compare it against. The tool name 'get_transcript' follows a clear verb_noun pattern.

Tool Count2/5

One tool is too few for a server named 'YouTube MCP Server', which suggests a broader scope. A single transcript retrieval tool feels incomplete for typical YouTube operations like searching videos, listing playlists, or managing uploads.

Completeness1/5

The tool surface is severely incomplete for a YouTube server. It only covers transcript retrieval, with no tools for core YouTube functionalities such as video search, metadata retrieval, playlist management, or user interactions, leaving significant gaps.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

Appeared in Searches