---
title: "Discussion on \"How to Scrape TikTok with Node.js: Comment Threads, Reply Pages, Captions, and a Comment-Section Report (2026)\" | Hashnode"
description: "Discussion on \"How to Scrape TikTok with Node.js: Comment Threads, Reply Pages, Captions, and a Comment-Section Report (2026)\". Pull TikTok comment threads with plain Node.js: page past the false has_more stop, fetch reply threads, pseudonymize commenters, grab WebVTT captions and build a comment-section report, with real Nov 17, 2026 runs and TikTok's terms up front."
url: "https://hashnode.com/posts/how-to-scrape-tiktok-with-nodejs-comment-threads-reply-pages-captions-and-a-comment-section-report-2026/6afbd5a236234e4a102270e0"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [How to Scrape TikTok with Node.js: Comment Threads, Reply Pages, Captions, and a Comment-Section Report (2026)](https://viktorhendelmann.hashnode.dev/how-to-scrape-tiktok-with-nodejs-comment-threads-reply-pages-captions-and-a-comment-section-report-2026?utm_source=hashnode&utm_medium=feed)

Pull TikTok comment threads with plain Node.js: page past the false has\_more stop, fetch reply threads, pseudonymize commenters, grab WebVTT captions and build a comment-section report, with real Nov 17, 2026 runs and TikTok's terms up front.

2d ago · 27 min read · viktorhendelmann.hashnode.dev  
Read the article

-   [#nodejs](https://hashnode.com/tag/nodejs)
-   [#javascript](https://hashnode.com/tag/javascript)
-   [#webscraping](https://hashnode.com/tag/webscraping)
-   [#tiktok](https://hashnode.com/tag/tiktok)
-   [#tutorial](https://hashnode.com/tag/tutorial)

## Written by

[Viktor](https://hashnode.com/@viktorhendelmann)

@viktorhendelmann

  
[View profile](https://hashnode.com/@viktorhendelmann)

## Responses 1

Markdown supported · ⌘↵ to post

[OnlineProxy](https://hashnode.com/@onlineproxyio)

[22h ago](https://hashnode.com/posts/how-to-scrape-tiktok-with-nodejs-comment-threads-reply-pages-captions-and-a-comment-section-report-2026/6afbd5a236234e4a102270e0/comment/6afdd1f89e33d578183662bf)

the caption echo needs a caveat too. the track is the eng US machine translation the startsWith check picks, so you are really measuring translation echo.

the italy number is just the 24 page cap, the offset 2000 and 5000 claims don’t show up in the logs, and the rate limit evidence is thin, around 120 requests from one ip. the reply loop still trusts has\_more, which goes against your own lessons. for brand team readers tiktok api for business apparently has comment list and reply list endpoints with webhooks, but I couldn’t confirm them so check the developer docs first
