TubeZero

📖 Read the Documentation

A modular collection of JavaScript functions designed to scrape and reverse-engineer data from YouTube using its private internal API (InnerTube). This library is built to be ultra-lightweight, free of heavy external dependencies, and optimized for client-side environments (such as browser extensions or desktop applications).

---

⚖️ Comparison with Other Libraries

To ensure complete transparency, here is a comparison between TubeZero and the other main InnerTube libraries available on GitHub:

Feature/Property	TubeZero (This Project)	LuanRT/YouTube.js (youtubei.js)	SuspiciousLookingOwl/youtubei
Architecture	Lite, pure fetch client, zero external dependencies	Full-featured monolithic SDK mapping almost all renderers	Object-oriented TypeScript query client
Optimization	Client-side & Web-Extension optimized	Heavy backend/server-side oriented	Node.js oriented with class-based queries
Pros & Use Cases	Extremely lightweight, fast, no compilation required. Native extraction of SAPISID and SAPISIDHASH via Web Crypto API for authenticated requests.	Complete coverage of features (deciphering video download signatures, live chat, dedicated interfaces for Music/Studio/Kids).	Class-based abstractions, clean TypeScript interface for querying videos, channels, and playlists.
Limitations	Lacks video download deciphering, full catalog of renderer schemas, and dedicated YT Music/Studio/Kids interfaces.	Large bundle size, complex codebase, requires polyfills/build tools to run in client/extension frontends.	Requires compilation, coupled with Node.js dependencies, not designed for direct client-side integration.

---

⚠️ Important Note on CORS and Cookies

When this library is executed in a standard browser environment (e.g., on a standard website like myapp.com), requests to youtube.com/youtubei will be blocked due to browser CORS policies.

Recommended Execution Environments:

1. Browser Extensions: Optimal. By granting host permissions for https://*.youtube.com/* in manifest.json, the extension bypasses CORS and can read session cookies directly to compute the SAPISIDHASH for authenticated requests.
2. Desktop & Mobile Applications (Electron, Tauri, React Native): By controlling HTTP headers and disabling CORS, these applications can run the library natively.
3. Server-side Environments (Node.js) with a CORS Proxy: Requests can be routed through a reverse proxy that removes header restrictions.

---

📦 Installation (Package Structure)

To prepare the package for local usage or NPM publishing:

# Initialize the package if exported into a dedicated folder
npm init -y

Add "type": "module" in your package.json to enable loading of ES modules.

---

💻 Usage Examples

1. Initialization and Fetching Taste Data (History and Likes)

Retrieve the logged-in user's viewing history, liked videos, "Watch Later" list, or custom playlists.

import { scrapeTasteData } from './index.js';

// When run in a browser extension, SAPISID cookies are read automatically
const userFeeds = await scrapeTasteData(null, [
    { url: "https://www.youtube.com/playlist?list=PLtbcYJeD7QZ34kS_L8H-lV7J8xT6rU0k_" } // Optional custom playlist
]);

console.log("History:", userFeeds.historyEntries);
console.log("Likes:", userFeeds.likesEntries);
console.log("Watch Later:", userFeeds.wlEntries);
console.log("Custom Playlists:", userFeeds.customPlaylistsData);

2. Fetching Video Comments

Retrieve comments for a specific video with automatic pagination.

import { fetchCommentsFromYouTube } from './index.js';

const videoId = "dQw4w9WgXcQ";
const maxComments = 100;

const comments = await fetchCommentsFromYouTube(videoId, maxComments);
comments.forEach(c => {
    console.log(`[${c.author}] (${c.likeCount} likes): ${c.text}`);
});

3. Fetching Subtitles (Transcripts)

Download and parse video transcripts in the desired language with automatic fallbacks.

import { fetchSubtitlesFromYouTube } from './index.js';

const videoId = "dQw4w9WgXcQ";
const transcript = await fetchSubtitlesFromYouTube(videoId, "en"); // Searches for English, falls back to automatic captions or first available track

transcript.forEach(segment => {
    console.log(`[${segment.start}s - ${segment.duration}s]: ${segment.text}`);
});

---

🛡️ Legal Disclaimer

This tool is intended solely for educational, research, and study purposes regarding third-party web architectures. It is not associated, affiliated, sponsored, or endorsed by Google LLC or YouTube. The user assumes full civil and administrative liability arising from the use of these functions and any potential violation of YouTube's Terms of Service. The authors are not responsible for IP blocks, account bans, or other actions taken by YouTube as a result of using this library.