Brave Discussions Scraper

Scrapes forum and discussion results from Brave Search. Extracts source-backed threads from Reddit, Stack Exchange, and other community sites that Brave places in search results or inline discussion cards.

What this Actor collects

Each dataset item represents one Brave Discussions result from a public community source, with the thread URL, excerpt, community, dates, reply or vote data, and top-answer details when exposed.

• Uses the input limits and filters below to control the crawl.
• Stores source-backed fields defined by the 31-field dataset schema.
• Omits optional fields when the source does not expose a value instead of writing nulls or fabricated placeholders.

Use cases

• Community and audience research
• Question and answer monitoring
• Discussion-source enrichment

Input

Provide input in JSON. Fields marked required must be supplied; source-specific alternatives and constraints are described in the field text.

Field	Type	Required	Default	Description
query	string	Yes	—	The search query to find discussion/forum results on Brave Search.
maxItems	integer	No	30	Maximum results, from 1 to 100.
maxPages	integer	No	3	Maximum Brave pages, from 1 to 10.
country	string	No	"us"	Brave country code for localized results.
language	string	No	"en"	Brave language code.
maxRequestRetries	integer	No	2	Retries for temporary navigation, rate-limit, and browser failures (0–5).
navigationTimeoutSecs	integer	No	60	Navigation timeout, from 10 to 120 seconds.
requestTimeoutSecs	integer	No	90	Validation and extraction timeout, from 15 to 180 seconds.
headless	boolean	No	true	Run Firefox without showing a browser window.
proxyConfiguration	object	No	direct connection	Optional proxy settings; enable only a group authorized for your account.

Example input

{
  "query": "react vs vue",
  "maxItems": 10,
  "maxPages": 1,
  "country": "us",
  "language": "en",
  "proxyConfiguration": {
    "useApifyProxy": false
  },
  "maxRequestRetries": 2,
  "navigationTimeoutSecs": 45,
  "requestTimeoutSecs": 90,
  "headless": true
}

Output

The default dataset contains one item per discussion result. The following are the most useful fields; answer, reply, and vote fields appear only when Brave exposes them.

Field	Type	Description
position	integer	Position
title	string	Thread title
link	string	Thread URL
source	string	Source
community	string	Community
snippet	string	Question Snippet
answerCount	integer	Answer Count
voteCount	integer	Top Answer Votes
publishedAt	string	Published At
searchQuery	string	Search Query
scrapedAt	string	Scraped At
page	integer	Page
domain	string	Domain
threadId	string	Thread ID
displayUrl	string	Displayed URL
snippetHtml	string	Question HTML
publishedAtRaw	string	Published At Raw

Example dataset item

This compact example is taken from local Actor storage. Long text and nested collections are shortened for documentation only.

{
  "position": 1,
  "link": "https://www.reddit.com/r/reactjs/comments/10u17c7/what_does_react_do_better_than_vue_innately/",
  "source": "Reddit",
  "community": "reactjs",
  "snippet": "Things like ecosystem could be explained away by existing earlier. I am curious what is inately better about the React library over Vue.",
  "answerCount": 18,
  "voteCount": 235,
  "publishedAt": "2023-02-05T00:00:00.000Z",
  "searchQuery": "react vs vue",
  "scrapedAt": "2026-07-23T15:27:01.889Z",
  "page": 1,
  "domain": "reddit.com"
}

How it works

The Actor opens one localized Brave Search page at a time with Firefox, validates the HTTP status and HTML content, rejects verification/challenge pages, and extracts only web cards that contain Brave inline Q&A content or come from a recognized forum/discussion domain. Similar-thread links are normalized into separate records. URLs are deduplicated across pages, HTML excerpts are sanitized, and the hard item limit is enforced before storage.

Cost and performance

Browser startup and page rendering dominate compute cost. Start with maxItems: 5, maxPages: 1, and zero retries while testing. Concurrency is fixed at one to reduce blocking and preserve deterministic positions. Proxy traffic is optional and should be enabled only after direct access is blocked.

Limitations and responsible use

• Brave may show a human-verification page instead of results. The Actor fails transparently and stores no challenge content.
• Not every query has a Discussions module; ordinary results are included only when their domains clearly identify a forum or community source.
• Reply counts, votes, dates, excerpts, and top answers are optional because Brave does not expose them on every card.
• Respect source terms, copyright, privacy, and applicable law. Avoid collecting or republishing personal or sensitive discussion content without a valid basis.

FAQ

Why did the Actor return no records? The query may have no discussion results, or Brave may have challenged the route. Check the first run error before changing proxy settings.

Does the Actor visit each forum thread? No. It extracts only information Brave exposes in its result cards, so it is bounded and does not crawl linked communities.

For support, provide the run ID, sanitized input, locale, and first error message. Never include proxy credentials or private data.

Related Actors

• Brave AI Mode Scraper
• Brave Autocomplete Scraper
• Brave Goggles Scraper