Ready
Ready

Enter parameters and send a request. The task status and normalized results will appear here.

README

YouTube Comment Replies

YouTube Comment Replies are public responses nested beneath a top-level video comment, with reply text, author identity, visible likes, publication time, and parent or video context. These thread records support conversation reconstruction, response review, and public-discussion dataset preparation.

Features

  • Known-thread token scopeStart from the opaque reply continuation token attached to a compatible top-level comment; video URLs and comment IDs are not interchangeable inputs.
  • Reply and author recordsReturn reply IDs, public text, associated URLs, publication times, and normalized author or channel identity.
  • Thread and creator contextKeep visible likes, nested-reply counts, parent comment and video references, and channel-owner or creator-heart markers together.
  • Bounded continuation collectionFollow subsequent source continuation tokens, remove duplicate reply IDs, and stop at the selected limit, an exhausted thread, or the 100-page ceiling.

Parameters

ParametersRequiredDescription
continuation_tokenRequiredRequired non-empty reply token from extra.replies_continuation_token on a compatible top-level YouTube Comments result. It is not a video URL or comment ID.
results_limitOptionalMaximum public replies to collect from this thread. Defaults to 100 and accepts integers from 20 through 2000.

How to Use

Continue a known public comment thread as an asynchronous task, then read normalized reply records from the completed task.

  1. Run YouTube Comments for a public video URL and select a top-level comment whose result includes extra.replies_continuation_token.
  2. POST that opaque continuation_token and a results_limit from 20 through 2000 to /v1/youtube/comment-replies.
  3. Retain the returned task_id and poll GET /v1/tasks/{task_id} while the task is queued, running, or retrying.
  4. After success, read data.results.items and pass data.results.next_cursor as cursor on the task request while data.results.has_more is true.
REPLY_TOKEN='paste-extra.replies_continuation_token'

curl -X POST "https://api.socq.ai/v1/youtube/comment-replies" \
  -H "Authorization: Bearer $SOCQ_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"continuation_token\":\"${REPLY_TOKEN}\",\"results_limit\":100}"

# Poll GET /v1/tasks/{task_id}
# Read data.results.items and pass data.results.next_cursor as cursor while has_more is true

Best Use Cases

  • Known-thread reconstruction: Use reply IDs, parent_comment_id, video_id, URLs, and timestamps to reconnect saved responses with a selected top-level comment.
  • Reply-level audience review: Compare public reply text, author identity, visible likes, nested-reply counts, and creator interaction markers within one known thread.
  • Public discussion dataset preparation: Combine normalized reply text, authors, timestamps, and thread context into records for manual review or downstream text processing.

Pro Tips

  • Obtain continuation_token from extra.replies_continuation_token on the matching top-level YouTube Comments record so the reply thread and source video remain aligned.
  • Copy the continuation token as an opaque value; if it is missing, expired, or rejected, collect the top-level comments again instead of substituting a video URL or comment ID.
  • Keep the source continuation_token separate from the cursor returned by GET /v1/tasks/{task_id}; the first selects a YouTube reply thread, while the second pages through saved task results.
  • Set results_limit to the expected thread size, but treat it as a ceiling because public visibility and thread exhaustion can reduce the returned count.
  • Deduplicate saved records by reply ID and retain parent_comment_id, video_id, and extra.reply_token when joining replies back to their source comment.
  • Allow nullable author, timestamp, metric, and creator-marker fields, and escape public reply text before displaying it in another interface.

Use these APIs when you need a different type of public YouTube data.

  • YouTube Channel Videos APICollect regular public videos from channel URLs, excluding Shorts and live streams.
  • YouTube Comments APICollect public comments, authors, engagement, source videos, and conversation context.
  • YouTube Transcripts APIRetrieve available plain-text transcripts with optional language selection.
  • YouTube Channels APIRetrieve public channel identity, images, verification, and visible audience statistics.
  • YouTube Videos APIRetrieve metadata, thumbnails, duration, and engagement for known video or Short URLs.
  • YouTube Shorts APICollect public Shorts, thumbnails, duration, publication details, and engagement by channel.
  • YouTube Search APISearch public regular videos by keyword with supported metadata filters.
  • YouTube Channel Live Videos APICollect public broadcast records exposed on channel live surfaces, with source context and visible metrics.
  • YouTube Community Posts APICollect public Community-tab posts from channel URLs, including text, channel identity, timestamps, image references, and source context.
  • YouTube Hashtag Search APIFind public videos from hashtag feeds with all-content or Shorts-only collection.
  • YouTube Playlist Videos APICollect video identity, channel context, thumbnails, duration, and playlist provenance from public playlist URLs or IDs.

YouTube Comment Replies APIFAQ