YouTube Subtitle Acquisition Gateway

An Model Context Protocol (MCP) endpoint facilitating the extraction of complete caption sets and transcript data from specified YouTube content. This service offers direct programmatic entry to available video narration text.

Deployment via Smithery

To deploy the YouTube Subtitle Acquisition Gateway to your Claude Desktop environment using Smithery:

bash npx -y @smithery/cli install @kimtaeyoon83/mcp-server-youtube-transcript --client claude

Functional Modules

Provided Tools

get_transcript
Mechanism for pulling textual records from YouTube media items
Parameters:
- url (string, mandatory): The full locator link or unique identifier for the YouTube asset.
- lang (string, optional, preset: "en"): The linguistic identifier (e.g., 'fr', 'es') for the desired subtitle track.

Core Capabilities

Adaptability across various YouTube addressing schemes
Precision fetching based on specified linguistic preferences
Comprehensive data payloads accompanying the extracted text

Operational Setup

To integrate this gateway within Claude Desktop, incorporate the following configuration segment:

{ "mcpServers": { "youtube-transcript": { "command": "npx", "args": ["-y", "@kimtaeyoon83/mcp-server-youtube-transcript"] } } }

Installation via Command Utility

mcp-get A dedicated CLI utility for managing the setup and lifecycle of Model Context Protocol (MCP) services.

shell npx @michaellatman/mcp-get@latest install @kimtaeyoon83/mcp-server-youtube-transcript

Repository Catalog

awesome-mcp-servers A curated index of superior Model Context Protocol (MCP) endpoints.

Development Lifecycle

Necessary Preconditions

Runtime environment: Node.js version 18 or newer
Package manager: npm or yarn

Initialization Sequence

Acquire required dependencies: bash npm install

Compile the server application: bash npm run build

For iterative development with automatic recompilation: bash npm run watch

Verification

Execute automated checks: bash npm test

Diagnostic Assistance

As MCP services interface via standard input/output streams, tracing execution can be complex. We advise utilizing the dedicated MCP Inspector utility during development phases:

bash npm run inspector

Executing Benchmarks (Evals)

The evals package utilizes an integrated mcp client to run scripts within index.ts, negating the need for rebuilds between test iterations. Environment variables can be preloaded by prefixing the execution command. Complete documentation is accessible via https://www.mcpevals.io/docs.

bash OPENAI_API_KEY=your-key npx mcp-eval src/evals/evals.ts src/index.ts

Failure Management

The gateway incorporates comprehensive logic to address typical operational disruptions, including: - Malformed video locator strings or identifiers - Inaccessibility of requested caption tracks - Issues related to language code compliance - Network communication failures

Application Scenarios

Retrieving subtitles using the full video web address: typescript await server.callTool("get_transcript", { url: "https://www.youtube.com/watch?v=VIDEO_ID", lang: "en" });
Retrieving subtitles using only the unique video identifier: typescript await server.callTool("get_transcript", { url: "VIDEO_ID", lang: "ko" });
Procedure to obtain YouTube captions within the Claude Desktop Interface:

chat: https://youtu.be/ODaHJzOyVCQ?si=aXkJgso96Deri0aB Extract subtitles

Security Posture

The service guarantees: - Rigorous validation of all supplied parameters - Graceful management of errors returned by YouTube interfaces - Imposition of time limits for all subtitle acquisition attempts - Provision of explicit error messaging to aid debugging efforts