Mine TikTok comments for audience questions and objections
Pull the comments, and optionally the replies, from TikTok videos as JSON to find what viewers ask, doubt and complain about.
A TikTok video can hold thousands of comments, and the useful ones are buried: the questions nobody answered, the objection that keeps coming back, the feature people ask for. Reading them by hand does not scale past one video. Lovely Router collects them for you. Send video links, get one JSON object per comment, pay per comment from a prepaid balance. Run it from a script, a notebook or an AI agent with one API key.
Hand it to your AI agent
Copy this prompt into Claude Code, Cursor, Codex or any coding agent. It tells the agent where the docs are and how to use your API key.
Mine TikTok comments for audience questions and objections. Use the Lovely Router data tools: each returns JSON and is billed per result.
- Docs (read first): https://lovelyrouter.com/docs/tools/find.md and https://lovelyrouter.com/docs/tools/run.md
- Plan first: POST https://lovelyrouter.com/v1/tools/suggest with `{"task": "<my task in plain words>"}` (free). It returns the tools to run in order, each with an example `input`, the `input_required` fields, an estimated cost and a ready-made `run` request. Replace the example values with my task's; if a field is unclear, read the tool with GET https://lovelyrouter.com/v1/tools/<id>. To browse instead, POST https://lovelyrouter.com/v1/tools/discover searches the catalog.
- Show me the plan and its estimated cost (`total_est_cost_usd`) before running anything large.
- Run: POST https://lovelyrouter.com/v1/tools/<id>/run with `{"input": {...}, "max_results": N}`. A 200 has the finished run with `items`; a 202 means it is still running, so poll the `poll` path until `status` is `succeeded`, `failed`, `timed_out` or `stopped`.
- Only delivered results are billed. Keep max_results small on the first run, check the output, then scale up.
- If you support MCP, connect https://lovelyrouter.com/mcp instead (Streamable HTTP, same key; https://lovelyrouter.com/docs/tools/mcp.md) and use suggest_tools, get_tool and run_tool.
- Auth: read my API key from the LOVELY_API_KEY environment variable and send it as `Authorization: Bearer $LOVELY_API_KEY`. Never print or commit the key. If it is not set, ask me to create one at https://lovelyrouter.com/app/keys.
- Billing is prepaid: every call is charged to my balance, so test with a small request before running anything large.
- Every docs page is plain Markdown; the full index is https://lovelyrouter.com/llms.txt.
My task: <the inputs (links, keywords or accounts) and how many results you need>How it works
Pick the videos worth reading
Choose your own top posts, a competitor's launch video or a viral video in your niche. Paste the video links, one per entry.
Run the tool with a cap
Send the links to
POST /v1/tools/tiktok-comments/runwithmaxItemsin the input andmax_resultson the request. SetincludeRepliestotruewhen the threads under each comment matter.Sort the comments by theme
Hand the comments to a model and ask for questions, objections, requests and praise, each with counts and a few example quotes.
Get the comments of a TikTok video with one request
List TikTok comments takes video links in startUrls. Only post, video and photo links are accepted. maxItems caps how many comments come back, and includeReplies adds the replies under each comment:
curl https://lovelyrouter.com/v1/tools/tiktok-comments/run \
-H "Authorization: Bearer $LOVELY_API_KEY" \
-H "content-type: application/json" \
-d '{
"input": {
"startUrls": [
"https://www.tiktok.com/@billieeilish/video/7050551461734042926"
],
"maxItems": 20
},
"max_results": 20
}'Comments come back one object per comment, so a model can read them as a plain list. Keep the cap small while you test. A run still going after 60 seconds returns 202; Run a tool shows how to poll.
What each comment holds
| Field group | What you get |
|---|---|
| Comments | The comments on the video, found from the video link |
| Replies | Replies under each comment, when includeReplies is true |
The tool is described as returning comments on a video by URL. Check one small run to see exactly which fields your account gets before you build a pipeline on them.
Recipes
Run your five most-commented videos with a cap of 500 each, ask a model to extract every question, group the duplicates and rank them by how often they appear. The top ten become your next FAQ or video series.
After a product video, pull the comments with replies on and ask a model to list the reasons viewers hesitate: price, size, shipping, trust. Fix the top one in the product page.
Run the launch video at 9 am and again the next day. Compare the two to see whether sentiment moved and which new questions appeared.
Run one video from each account with the same cap and ask a model what each audience asks for that the other brand does not hear.
Search for a trend with tiktok-search, take the top video and read its comments for "where can I buy" and "does it come in" requests.
Prefer plain words? Connect the MCP server and tell Claude Code or Cursor: "Read the comments with replies on these three TikTok videos, group them into questions, complaints and requests, and give me the top five of each with example quotes. Set max_results to 300."
Pair it with other TikTok tools
Comments tell you what viewers think. Search, profile and transcript data tell you what was said and who said it:
Comments on a TikTok video by URL.
TikTok videos for keywords or hashtags: caption, views, likes, shares, author, music.
A TikTok account's videos with the author's follower and like counts.
Spoken-word transcripts of TikTok videos by URL.
Comments on YouTube videos or Shorts: text, likes, replies count, author, date. Sorted by top or latest.
tiktok-searchfinds the videos worth reading by keyword or hashtag.tiktok-profilelists an account's videos, so you can pick the ones to read.tiktok-transcriptgives the spoken words, to read the video next to its comments.youtube-commentsdoes the same job on YouTube.
For brand-wide listening across platforms, see social listening.
What you pay
You pay per comment: starting a run holds its maximum cost, the per-call fee plus the per-comment fee times max_results, and the run settles to what it returned. The rest of the hold goes back to your balance. A run that fails or times out pays only for the comments it delivered, and a run refused at the start costs nothing.
Reading 100 comments is billed as 100 comments. There is no subscription and no monthly minimum: top up from $5, and one balance pays for tools and models. The card above shows today's price; pricing lists every tool.
Frequently asked questions
Can I get replies as well as top-level comments?+
Yes. Set includeReplies to true in the input and the tool also fetches the replies under each comment it finds. Replies count as results, so raise your cap accordingly.
What links does it accept?+
TikTok post, video and photo links. Profile, search and hashtag links are not supported by this tool; use tiktok-search or tiktok-profile for those.
How many comments can one run return?+
The cap is set with maxItems in the input, and the tool's maximum is 500 comments per run. For a bigger sample, start several runs on different videos.
Does the data include who wrote each comment?+
The tool is described as returning the comments on a video. Run a small test to see which fields your account receives, and treat any usernames you store as personal data.
Do I need a subscription?+
No. You add credit to a prepaid balance and each run draws it down. Failed runs pay only for the comments they delivered, and nothing renews automatically.
Looking for more? See the rest of the tool catalog and the pricing page.