YouTube Comments Scraper & API Client for JavaScript / TypeScript

The official JavaScript & TypeScript SDK for the FetchLayer YouTube API.

Scrape YouTube comments. Pull reply threads. Extract video transcripts. Get likes, authors, and dates. All as clean JSON from a single API — no Google Cloud project, no OAuth, no HTML parsing.

Open-source SDK, hosted API. This package is MIT-licensed. The API requires a free FetchLayer API key — no subscriptions, no contracts, no lock-in. Pay only for what you use.

Install

npm install @fetchlayer/youtube

pnpm add @fetchlayer/youtube

yarn add @fetchlayer/youtube

Get your API key

1. Go to fetchlayer.dev
2. Sign in with your email
3. Copy your API key from the dashboard

That's it. No credit card. No trial expiration. You get free requests to start, then pay-as-you-go when you need more.

Quick start

import { FetchLayerYoutube } from "@fetchlayer/youtube";

const youtube = new FetchLayerYoutube({
  apiKey: process.env.FETCHLAYER_API_KEY!,
});

// Scrape comments from any YouTube video
const comments = await youtube.getComments({
  videoUrl: "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  sort: "top",
  pages: 2,
});

console.log(comments);

That's 4 lines to get structured YouTube comment data. No Google Cloud setup, no OAuth dance, no broken scrapers.

Why use this instead of scraping YouTube yourself

| Problem | FetchLayer | | --- | --- | | YouTube blocks your IP | We handle rotation and resilience | | HTML structure changes break your parser | You get stable JSON | | YouTube Data API needs a Google Cloud project + quota | Simple Bearer token, no quota | | Maintaining scraping infra is a full-time job | npm install and done | | You need comments, replies, AND transcripts | Three endpoints, one SDK |

All methods

const youtube = new FetchLayerYoutube({ apiKey });

| Method | What it does | | --- | --- | | youtube.getComments(params) | Scrape top-level comments from a video | | youtube.getCommentReplies(params) | Fetch replies to a specific comment | | youtube.getTranscript(params) | Extract video transcript with timestamps |

Examples

Scrape top comments from a video

const response = await youtube.getComments({
  videoUrl: "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  sort: "top", // "top" (most liked) or "newest"
  pages: 3,    // scroll for more comments
});

for (const comment of response.comments ?? []) {
  console.log(`${comment.username}: ${comment.comment}`);
  console.log(`  ${comment.likes} likes · ${comment.date}`);
}

Get the latest discussion

const response = await youtube.getComments({
  videoUrl: "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  sort: "newest",
  pages: 5,
});

Fetch replies to a specific comment

const thread = await youtube.getCommentReplies({
  videoUrl: "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  commentId: "Ugz...",
  depth: 2,
});

console.log(thread.comment);
for (const reply of thread.nestedReplies ?? []) {
  console.log(`  ↳ @${reply.username}: ${reply.comment}`);
}

Extract a video transcript

const transcript = await youtube.getTranscript({
  videoUrl: "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  language: "en",
});

console.log(transcript.text);
for (const segment of transcript.segments ?? []) {
  console.log(`[${segment.start}s] ${segment.text}`);
}

Error handling

import { FetchLayerYoutube, FetchLayerError } from "@fetchlayer/youtube";

try {
  const comments = await youtube.getComments({ videoUrl: "https://..." });
} catch (err) {
  if (err instanceof FetchLayerError) {
    console.error(err.status); // 401, 429, etc.
    console.error(err.message); // Human-readable error from the API
  }
}

Configuration

const youtube = new FetchLayerYoutube({
  apiKey: "your-key",     // Required
  baseUrl: "...",         // Default: https://api.fetchlayer.dev/youtube
  timeoutMs: 120000,      // Default: 120s
  fetch: customFetch,    // Default: global fetch (Node 18+)
});

What this package is

• Zero dependencies — uses native fetch, nothing else
• Dual build — ESM + CommonJS, works everywhere
• Fully typed — TypeScript declarations included
• Tiny — under 6 KB bundled
• Open source — MIT license, contributions welcome

How it works

You call FetchLayer → FetchLayer returns structured YouTube comment data as JSON → you use it however you want.

No YouTube account needed. No Google Cloud project. No scraping infrastructure on your end.

Pricing

| | | | --- | --- | | Free tier | Included with every account — start building immediately | | Pay-as-you-go | $1.99 per 1,000 requests. Credits never expire. | | Subscriptions | Available for heavy use — predictable pricing at scale | | Rate limits | None |

No credit card required to start. No monthly commitment. No overages. Get your API key →

Use cases

• Audience research — find what viewers ask, praise, and complain about in the comments
• Content research — discover recurring questions and content requests for your next video
• Competitive intelligence — monitor comments on competitor videos and launches
• Sentiment analysis — pull comments and run NLP on real viewer opinions
• Community management — track feedback and engagement across your channel
• AI agents — give your LLM real-time YouTube context via FetchLayer's MCP server

Links

| | | | --- | --- | | Get API key | fetchlayer.dev/signin | | Product page | fetchlayer.dev/youtube-comments-api | | Homepage | fetchlayer.dev | | npm | @fetchlayer/youtube | | Issues | GitHub Issues |

License

MIT — see LICENSE.