Twitch Live Streams Scraper

Discover live Twitch channels by category. Export twelve columns per live channel - channel URL, login and display name, stream title, category, tags, viewer count, verified badge, avatar, preview thumbnail and live status - without an API key. Results are returned as clean, typed dataset rows suitable for CSV, JSON, automation, and AI-agent workflows.

What data can I extract?

Every record carries twelve columns. The dataset schema documents each one, with a real example value, before a run starts.

Field	What it holds
stream_title	Stream title as the streamer wrote it, original capitalization
channel_name	Lowercase channel login used in the Twitch URL
channel_display_name	Public display name, capitalization and non-Latin characters preserved
channel_url	Canonical twitch.tv/<channel> URL
stream_category	Game or category the stream is listed under
stream_tags	Card tags, comma-separated (language, content and community tags)
viewer_count	Point-in-time viewer count label, for example 1.7k viewers
verified_badge	Verified when Twitch shows the badge, otherwise empty
profile_image_url	Channel avatar image URL
thumbnail_url	Live preview thumbnail URL
live_status	Point-in-time live-state label
stream_label	Accessibility label combining title and channel

stream_category is filled when you scrape a multi-category listing URL such as twitch.tv/directory/all through start_urls. On a single-category page the category is your input, so Twitch omits it from the card and the column is empty.

Input example

{
  "search_query": "just-chatting",
  "max_items": 10,
  "max_pages": 1
}

max_items is a hard output and billing ceiling. Source fields may be null when the upstream public record does not provide them.

Output example

Each dataset item is one live channel, emitted as typed JSON properties and exportable as CSV, Excel, XML, or RSS.

{
  "stream_label": "subtember! - fri- games - reacts - no order - lowtiergod",
  "channel_url": "https://www.twitch.tv/lowtiergod",
  "channel_name": "lowtiergod",
  "viewer_count": "1.7k viewers",
  "thumbnail_url": "https://static-cdn.jtvnw.net/previews-ttv/live_user_lowtiergod-640x360.jpg",
  "live_status": "live",
  "stream_title": "SUBTEMBER! - FRI- GAMES - REACTS - NO ORDER",
  "channel_display_name": "LowTierGod",
  "verified_badge": "Verified",
  "stream_tags": "justchatting, drops, DropsEnabled",
  "profile_image_url": "https://static-cdn.jtvnw.net/jtv_user_pictures/f5a3e1e9-4acb-4b46-a807-9c27dfeb54f9-profile_image-50x50.jpeg",
  "stream_category": null
}

Common use cases

• Lead, market, compliance, and competitive research
• Scheduled monitoring and change detection
• CSV/JSON enrichment pipelines
• AI-agent retrieval and workflow automation

Pricing and cost control

Output is billed per result at $0.003 per result (about $3.00 per 1,000 results), plus a $0.001 Actor-start charge billed once per gigabyte of memory at run start. Use max_items to cap both output volume and charges. The price shown on the Apify Store listing is authoritative.

Use with AI agents and MCP

Apify's MCP server can discover and call this Actor. Example intent:

Find 50 currently live Twitch channels in the just-chatting category.

Use the JSON input above as the tool arguments, changing filters and max_items as needed. The strict input schema rejects unknown properties, and the typed dataset schema describes the returned tool data.

Reliability

Runs use bounded retries, deduplication, hard result limits, and explicit failure handling. Browser-based sources are monitored by contract canaries and staged selector repair; official-API sources use deterministic contract canaries rather than DOM repair.

Limitations and responsible use

• Publicly available source data only; coverage and field availability depend on the upstream source.
• A successful search can legitimately return no matching records.
• Point-in-time fields can change after collection.
• Use the data in accordance with applicable law and source terms.
• Do not use results for prohibited eligibility, discrimination, or sensitive-person profiling decisions.

FAQ

What search phrases does this Actor support?

The input schema lists exact accepted fields and examples. Primary discovery terms: Twitch scraper, live stream search, Twitch channel discovery, stream viewer counts.

Can I control cost?

Yes. Set max_items; it is enforced as the maximum number of output records.

Can an AI agent call it?

Yes. Use Apify MCP search-actors, inspect this Actor's schema, then call it with the same JSON accepted in Console.

Does it require a Twitch account or API credentials?

No external API key or Twitch account is required. The Actor queries live channel listings directly.

Changes

See ./CHANGELOG.md for contract and maintenance updates.