Twitch Clips Scraper

What does Twitch Clips Scraper do?

This Actor extracts public clip-card metadata from one Twitch game category or channel page. It uses bounded HTTPS HTML requests to official Twitch hosts, follows only official redirects, deduplicates canonical clip URLs, and preserves public titles, channels, games, views, duration, age, and thumbnails. It does not use browser automation, private APIs, credentials, or proxy-bypass techniques.

Twitch may return a page without rendered clip cards or block a request. Those outcomes become minimal four-field diagnostics and are summarized in OUTPUT; no placeholder clip is emitted.

Why use it?

Use it for small public clip inventories, category checks, channel highlights, and scheduled public-page monitoring through Apify's API, scheduling, monitoring, and integrations. Twitch does not expose a verified public overall-clips directory, so provide exactly one gameName or channelName.

How to run

1. Open the Actor's Input tab.
2. Provide either a public game/category name or a Twitch channel login.
3. Choose the time range and bounded item/response/request controls.
4. Run the Actor and inspect normal clip rows, diagnostics, and OUTPUT.

Input

Field	Required	Description
gameName	One of	Public Twitch game/category; mutually exclusive with channelName.
channelName	One of	Public Twitch channel login; @ is accepted.
timeRange	No	24h, 7d, 30d, or all; default 7d.
maxItems	No	Maximum unique clips, from 1 to 500; default 50.
timeoutMs	No	Request timeout from 5000 to 120000 ms.
maxResponseBytes	No	Response limit from 100000 to 10000000 bytes.
maxRequestRetries	No	Bounded retries from 0 to 2; default 0.
requestDelayMs	No	Polite delay before the request, maximum 10000.
proxyConfiguration	No	Direct by default; optional Apify Proxy routing.
fixtureFile	No	Actor-relative JSON fixture for deterministic local QA.

Example input:

{
  "gameName": "Minecraft",
  "timeRange": "7d",
  "maxItems": 10,
  "timeoutMs": 30000,
  "maxRequestRetries": 0,
  "proxyConfiguration": { "useApifyProxy": false }
}

Output

Normal rows use recordType: "clip" and include canonical clip URL/slug, title, public channel and creator fields, game, views, duration, age, thumbnail, query/source provenance, and timestamp. Transport, response-size, extraction-stage, and run-summary fields stay out of dataset rows.

Example normal output:

{
  "recordType": "clip",
  "position": 1,
  "clipTitle": "Speedrun to the Nether",
  "clipUrl": "https://www.twitch.tv/pixelpilot/clip/AbC_123",
  "clipSlug": "AbC_123",
  "channelName": "PixelPilot",
  "gameName": "Minecraft",
  "viewCount": 12400,
  "duration": "1:23",
  "timeRange": "7d",
  "queryUrl": "https://www.twitch.tv/directory/category/minecraft/clips?range=7d",
  "scrapedAt": "2026-09-08T00:00:00.000Z"
}

Example diagnostic:

{
  "url": "https://www.twitch.tv/directory/category/minecraft/clips?range=7d",
  "error": "The public Twitch page returned no extractable clip cards. No placeholder clip was emitted.",
  "errorCode": "NO_PUBLIC_CLIP_DATA",
  "scrapedAt": "2026-09-08T00:00:00.000Z"
}

OUTPUT contains status, query, time range, transport, extracted/stored clip counts, diagnostic count, data availability, generator provenance, and completion time. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Tips, FAQ, and responsible use

Use fixture mode while changing extraction logic. BLOCKED, TIMEOUT, HTTP_ERROR, REQUEST_FAILED, and NO_PUBLIC_CLIP_DATA reflect the public response observed by this run. Missing title, views, duration, or thumbnail fields are omitted rather than invented.

This Actor is not affiliated with Twitch. Follow Twitch terms, robots guidance, rate limits, privacy rules, and applicable law. Use only public information and do not use this Actor to evade access controls.