Web scrapers

Comment Scraper API

Comments are where the opinion actually lives, and they are the hardest part of a page to reach. Platforms load them lazily, hide them behind pagination tokens, and nest replies several levels deep. These endpoints flatten all of that: comments come back with author, text, like count and timestamp, replies come back through their own call keyed on the parent comment so the thread tree can be rebuilt exactly, and a cursor carries you through threads that run to thousands.
Get your API keyEvery scraper
  • 9 endpoints
  • 5 platforms
  • from $1.49 per 1,000 calls
  • One key, one JSON shape

Comment Scraper API

9 endpoints across 5 platforms.

Comment threads, nested replies and reviews, fully paginated.

TikTok Video CommentsPull every comment on a TikTok video, page by page, with author, text, like count, reply count and timestamp.$1.49 per 1,000TikTok Comment RepliesPull the full nested reply thread under any TikTok comment, so you can rebuild the whole conversation, not just the top level.$1.49 per 1,000TikTok Shop Product ReviewsPull the reviews on any TikTok Shop product, page by page, with star rating, review text, reviewer and date.$1.49 per 1,000Facebook Post CommentsPull every comment on a Facebook post, page by page, with author, text, reaction count, reply count and timestamp.$1.49 per 1,000Facebook Comment RepliesPull the full nested reply thread under any Facebook comment, so you can rebuild the conversation rather than just the top level.$1.49 per 1,000YouTube Video CommentsPull every comment on a YouTube video, page by page, with author, text, like count, reply count and timestamp.$1.49 per 1,000YouTube Comment RepliesPull the full reply thread under any YouTube comment, with authors, like counts and timestamps at every level.$1.49 per 1,000Instagram Post CommentsPull every comment on an Instagram post, page by page, with author, text, like count, reply count and timestamp.$1.49 per 1,000Reddit Post CommentsPull the comment tree under any Reddit post, with author, body, score and nesting depth on every comment.$1.49 per 1,000

In every response

The same JSON, whichever platform you ask.

  • comments[] Each comment with author, text, like count and timestamp.
  • author Commenter handle, ID and avatar.
  • likeCount / replyCount Engagement on each individual comment.
  • replies[] / parentId / depth On the reply endpoints, the nested thread with its nesting level.
  • cursor Pagination token to pull the next page of a long thread.
curl "https://hproxy.com/api/v1/scrape/youtube/comment-replies?continuationToken=value" \
  -H "X-API-Key: hpx_your_key_here"
Response · YouTube Comment Replies200 OK · JSON
{
  "parentId": "72901...",
  "replies": [
    { "id": "72944...", "author": "@user", "text": "same thing happened to me",
      "depth": 1, "likeCount": 12, "createdAt": "2026-07-14T09:31:00Z" }
  ],
  "cursor": "eyJvZmZzZXQiOjIwfQ"
}

What people build with it

Where comment scraper api earns its keep.

  • Run sentiment analysis on what people say under a brand's posts rather than about them.
  • Mine support questions and objections straight out of comment threads.
  • Watch a competitor's comment sections for the complaints their customers raise.
  • Pull product reviews with their ratings for pricing and positioning research.

Comment Scraper API questions

Answered before you ask.

The scraper docs

Top-level comments come from the comment endpoint. Nested replies are their own call, keyed on the parent comment ID, because a thread can be far larger than the post it hangs off and merging them would make one call unbounded.

Yes. The order is part of the data, because it reflects what a real visitor sees first. Where an endpoint exposes a sort mode, it is a documented parameter rather than something we decide for you.

As many as the platform serves publicly. Each call returns a page plus a cursor, and you keep going until the cursor comes back empty. Billing is per call, so depth costs pages, not a subscription tier.

You get the handle, the ID and the avatar on every comment, which is enough to group by author or to look the account up through the profile endpoint when you need the full picture.

Docs and AI

Connect your scraper to an AI, any time.

The whole Scraper API is published as markdown, so an assistant reads every endpoint, every parameter and a real response without a key and writes your integration. And because a scrape is one GET that answers in JSON, an agent can run the scraper itself, on its own schedule, with your key.

Read the scraper docsSet up your assistant

Claude, ChatGPT, Cursor, Gemini, Copilot, Windsurf, Perplexity, Ollama and any assistant that speaks MCP.

The API as markdown

Paste /llms/scraper-api.txt into Claude, ChatGPT, Cursor or Copilot: every endpoint and the whole catalog, in one file. No key and no account to read it.

Open the markdown

An agent can call it

One GET with your key returns structured JSON, so an agent scrapes on its own schedule. No browser to drive, no parser to keep alive between runs.

See the endpoints

MCP server and CLI

Built by us for the free tools and hosted at mcp.hproxy.com: the proxy list, the checker and the IP lookup as tools in Claude, ChatGPT, Cursor or any assistant that speaks MCP. Nothing to install, no key.

Add it to your assistant

API docs

Every scraper endpoint documented end to end, with the parameters it takes and the response it really returns.

Read the docs

Other kinds of data

The catalog, by what you need rather than by platform.

Every platform

The same key works on all 23.

The whole catalog

More on comment scraper api

What comments covers

9 endpoints across 5 platforms: TikTok Video Comments, TikTok Comment Replies, TikTok Shop Product Reviews, Facebook Post Comments, Facebook Comment Replies, YouTube Video Comments, YouTube Comment Replies, Instagram Post Comments and Reddit Post Comments. They are separate scrapers behind one API key, and they answer in the same JSON shape, so the code that reads one reads all of them.

Why the shape matters

Comment threads, nested replies and reviews, fully paginated. Working platform by platform means a second integration for every network you add. Here the field names, the pagination and the error codes are the same wherever the data came from, so adding another platform is a change of URL rather than a rewrite.

What it costs

Calls in this group start at $1.49 per 1,000 and are billed from one balance, per call, with no subscription. Blocked, errored and timed-out calls are not charged, and each endpoint prices itself: a heavier lookup costs more than a simple one, and the page for each endpoint prints its own rate.

Ready when you are.

Your dashboard is ten seconds away. No sales call, no subscription, no minimum.

Get startedSee every scraperTalk to us