tiktok-api-unofficial (TypeScript)

Unofficial TikTok API library (TypeScript/Node): scrape video and user profile metadata, and optionally download videos, slides, and audio. No official API key required.

Install

From npm:

npm install tiktok-api-unofficial

From GitHub:

git clone https://github.com/8dazo/tiktok_scrapper.git
cd tiktok_scrapper/tiktok-api-ts
npm install
npm run build

Requirements: Node.js 18+ (for native fetch). If you see Could not locate the bindings file for better-sqlite3, run:

npm rebuild better-sqlite3

Usage

import { TTContentScraper } from "tiktok-api-unofficial";

const scraper = new TTContentScraper({
  waitTime: 0.35,
  outputDir: "data/",
  progressDbPath: "progress_tracking/scraping_progress.db",
  clearConsole: false,
});

// Add video IDs (from TikTok URL) and/or usernames
scraper.addObjects(["7398323154424171806"], "batch1", "content");
scraper.addObjects(["tiktok", "tagesschau"], "users", "user");

// Scrape: metadata only, or with media (mp4/jpeg/mp3)
await scraper.scrapePending({ scrapeFiles: false });   // metadata only
// await scraper.scrapePending({ scrapeFiles: true }); // also download videos/slides/audio

• Content: use type: "content" and video IDs from the video URL.
• Users: use type: "user" and usernames (with or without @).

Output is written under outputDir:

• content_metadata/<video_id>.json – video/post metadata
• user_metadata/<username>.json – user profile metadata
• content_files/ – mp4, jpeg, mp3 when scrapeFiles: true

Progress is stored in the SQLite database at progressDbPath (pending/completed/error per ID).

Test

npm test

Runs a metadata-only scrape of 1 video + 1 user and checks output files.

API (TypeScript)

• TTContentScraper – Main scraper (extends ObjectTracker). Options: waitTime, outputDir, progressDbPath, clearConsole.
• ObjectTracker – SQLite progress DB: addObject, addObjects, markCompleted, markError, getPendingObjects, getStats, getObjectStatus, getErrorObjects, resetErrorsToPending, resetAllToPending, clearAllData, close.
• BaseScraper – Low-level: scrapeMetadata(videoId), scrapeUser(username), scrapeBinaries(links).
• filterTiktokData(raw) – Normalize raw TikTok itemStruct into typed metadata.
• ObjectStatus – Enum: PENDING, COMPLETED, ERROR, RETRY.

Publish to npm

1. Log in: npm login
2. From tiktok-api-ts: npm publish (or npm publish --access public for a scoped package like @your-scope/tiktok-api-unofficial).

If the name tiktok-api-unofficial is taken, set a different name in package.json (e.g. tiktok-scraper-unofficial or @your-username/tiktok-api-unofficial).

Push to GitHub

1. Create a new repository on GitHub (e.g. tiktok_scrapper).
2. In your project root:

   git remote add origin https://github.com/8dazo/tiktok_scrapper.git
   git push -u origin main

   (Use master instead of main if your default branch is master.)

License

MIT.