DuckDuckGo Videos Scraper

Collect structured public video-search results from DuckDuckGo without launching a browser. The Actor obtains DuckDuckGo's public search token, requests the video JSON feed, validates each media URL, removes duplicates, and stores only verified records.

Input

• query (required): search phrase, up to 120 characters.
• maxItems: 1–150 results; default 50.
• maxPages: 1–3 result pages. When omitted, it is derived from maxItems.
• region: DuckDuckGo region code such as us-en or uk-en.
• timeRange: any time, past day, week, or month.
• duration: any, short, medium, or long.
• resolution: any, high, or standard definition.
• proxyConfiguration: optional Apify or custom proxy. Direct requests are the default.

Example:

{
  "query": "openai",
  "maxItems": 20,
  "maxPages": 1,
  "region": "us-en",
  "duration": "short"
}

Output

Each dataset item includes a stable source identifier, title, canonical videoUrl, source domain, query and page provenance, and scrape timestamp. When the source supplies them, records also include the embed and thumbnail URLs, duration, publisher, channel, provider, description, view count, and publication date. Optional unavailable fields are omitted rather than filled with placeholders.

The OUTPUT key-value-store record contains the accepted and unique item counts, query, and completion time. The run fails clearly if DuckDuckGo rejects the request, changes its response contract, or produces no verifiable results.

Notes

Search results and metadata belong to their respective publishers. Availability varies by query, region, filters, and DuckDuckGo's current public response. Use the data responsibly and comply with applicable terms and laws.