@tubeagentkit/langchain-getyoutubetranscript

YouTube transcript tools and a document loader for LangChain.js, powered by the GetYouTubeTranscript API. Give agents the transcript of any YouTube video (optionally with [m:ss] timestamps) and YouTube search, or load transcripts into RAG pipelines.

The API fetches transcripts on its own servers, so it works on serverless platforms and cloud servers without proxies, and without RequestBlocked / IpBlocked errors.

Python version: langchain-getyoutubetranscript.

Install

npm install @tubeagentkit/langchain-getyoutubetranscript @langchain/core zod

Get an API key at getyoutubetranscript.com/dashboard (free tier included):

GETYOUTUBETRANSCRIPT_API_KEY=sk_live_...

Or pass { apiKey } to any tool or the loader.

Tools

| Factory | Tool name | What it does | | --- | --- | --- | | youtubeTranscriptTool() | youtube_transcript | Transcript of a video (URL or ID) with title and channel. timestamps: true gives [m:ss] lines. | | youtubeSearchTool() | youtube_search | Search YouTube for videos or channels. JSON with a continuation_token for the next page. |

import { youtubeTranscriptTool } from '@tubeagentkit/langchain-getyoutubetranscript';

const tool = youtubeTranscriptTool();
console.log(await tool.invoke({ video: 'https://youtu.be/jNQXAC9IVRw', timestamps: true }));
// Title: Me at the zoo
// Channel: jawed
// Language: en
//
// [0:01] All right, so here we are, in front of the elephants

API errors (no captions, invalid video, out of credits) come back to the agent as Error: ... text instead of failing the run.

With an agent

import { createAgent } from 'langchain';
import { youtubeSearchTool, youtubeTranscriptTool } from '@tubeagentkit/langchain-getyoutubetranscript';

const agent = createAgent({
  model: 'openai:gpt-5-mini',
  tools: [youtubeSearchTool(), youtubeTranscriptTool()],
});

await agent.invoke({
  messages: [{ role: 'user', content: 'Find a short talk on transformers and summarize it with timestamps.' }],
});

Document loader

import { GetYouTubeTranscriptLoader } from '@tubeagentkit/langchain-getyoutubetranscript';

const docs = await new GetYouTubeTranscriptLoader(['https://youtu.be/jNQXAC9IVRw', '5e37ZT3SQbk']).load();
console.log(docs[0].metadata);
// { source: 'https://www.youtube.com/watch?v=jNQXAC9IVRw', videoId: 'jNQXAC9IVRw', title: 'Me at the zoo',
//   authorName: 'jawed', languageCode: 'en', wordCount: 39 }

Options: language, timestamps (pageContent becomes [m:ss] lines and metadata.segments holds { start, duration, text } per caption line), apiKey.

Pricing

Each transcript or search request uses one credit from your GetYouTubeTranscript account. Failed requests are not charged.

Links

• GetYouTubeTranscript API docs
• Node.js SDK (this package is built on it)
• License: MIT