Start with a keyword, discover public TikTok videos by relevance, recency, or likes, and focus the research set with a publication window.
Try TikTok Search API · Get an API key · Documentation · All SocQ examples
- Market and category scans: Use product, category, and use-case terms to map the TikTok content landscape and creator sample.
- Event momentum tracking: Combine date with a day or week window to follow recent content around launches, events, and news moments.
- High-like creative research: Order by likes and study recurring hooks, captions, music choices, and hashtag combinations.
- Brand mention monitoring: Run brand and product queries on a schedule and save snapshots for content and communications teams.
sort_byacceptsrelevance,date, orlikes.published_withinacceptsday,week,month,three_months, orsix_months.results_limitdefaults to 100.- Search rankings are dynamic;
extra.search_queryis included when available.
All requests use the shared asynchronous flow:
submit -> task_id -> poll task -> read every cursor page -> save results
cp .env.example .env
export SOCQ_API_KEY="your-api-key"Run the complete Node.js workflow:
cd node
npm startRun the complete Python workflow:
python3 -m pip install -r python/requirements.txt
python3 python/main.pyBoth examples load payload.example.json, retry transient API responses, wait
for task completion, read every cursor page, and save a filtered public-video discovery dataset for topic and creator research to
output/results.json.
Never expose SOCQ_API_KEY in browser code, mobile apps, public repositories,
screenshots, fixtures, or logs.
POST https://api.socq.ai/v1/tiktok/search
Authorization: Bearer <SOCQ_API_KEY>
Content-Type: application/json{
"query": "python tutorial",
"results_limit": 50,
"sort_by": "relevance",
"published_within": "month"
}The submit response contains data.task_id. Poll the task endpoint until
data.status becomes succeeded or failed, then continue with
data.results.next_cursor while data.results.has_more is true.
The Node.js and Python programs implement the production-shaped happy path:
- Load and validate configuration.
- Submit the endpoint-specific payload.
- Retry rate-pressure and transient server responses with bounded backoff.
- Poll the asynchronous task with a ten-minute application timeout.
- Stop cleanly on a failed task and surface the public error message.
- Read all cursor pages instead of silently returning only the first page.
- Write a stable JSON artifact containing task metadata and normalized records.
Use the synthetic files in fixtures/ for tests and documentation. They do
not contain customer, account, or production data.
See docs/production-notes.md for validation,
retry, timeout, pagination, deduplication, logging, and endpoint-specific
guidance.
- Use only publicly accessible TikTok profiles, videos, comments, hashtags, search results, and fields supported by the selected endpoint.
- Do not use the examples to access private profiles, restricted content, login-only surfaces, or authentication controls.
- SocQ is not an official API of the represented social platform and is not affiliated with or endorsed by that platform.
- Before production use, assess the laws, platform terms, privacy obligations, and retention requirements that apply to your organization and use case.
- Collect only the fields needed for a defined purpose, restrict access, set retention periods, and support correction or deletion workflows where required.
- Platform names and trademarks belong to their respective owners.
This section describes the public-data boundary; it is not legal advice or a guarantee that every use case is permitted in every jurisdiction.
| Path | Purpose |
|---|---|
curl/request.md |
Copy-paste submit, poll, and pagination requests |
node/index.mjs |
Complete Node.js workflow |
python/main.py |
Complete Python workflow |
payload.example.json |
Safe endpoint-specific request body |
fixtures/ |
Synthetic submit and task response shapes |
docs/production-notes.md |
Production integration guidance |