Quora Keyword Search & Author Discovery Scraper

Collect public Quora questions, full answers, author profiles, and topics without cookies or logins. Paste question, answer, profile, or topic URLs, or use best-effort keyword discovery with related-question expansion. The default path is HTTP; a residential browser fallback is automatic only after Quora challenges the direct, datacenter, and residential HTTP paths.

What this actor does

• Three modes: urls follows pasted Quora links, search discovers indexed question links and expands related questions, authors builds author profiles from profile or question URLs.
• Full question context: title, URL, topics, follower count, answer count, related questions, and the explicit unanswered state.
• Full answers: answer text, permalink, author name and profile link, credential, verified badge, upvotes, views, comment count, share count, exact creation date, AI label, and related-answer provenance.
• Author profiles: followers, following, answers, questions, posts, content views, join date, credential, location, employment, education, and expertise topics.
• Honest discovery: Quora's native search page is gated for anonymous visitors, so keyword mode uses a public web index and reports a clear status row when no indexed matches exist. Related-question expansion is the reliable discovery path.
• HTTP first, browser last resort: every page is fetched over HTTP first. Only if Quora challenges the direct, datacenter, and residential HTTP paths does the actor render that page in a real Chromium session through a residential connection. The same parser is used, so the output shape does not change.
• Leaves empty fields out of each row instead of writing guesses or nulls.

Output per question

• question_id, title, question_title, url, question_url, canonical_url
• topics[], topic_urls[]
• answer_count, answer_count_text, answer_count_scraped, last_answered_at
• follow_count, follow_count_text, follower_count
• related_questions[], related_questions_count
• question_language, answer_pagination_state
• discovery_source, source_query, parent_question_url
• access_state, scrape_timestamp, record_id

Output per answer

• answer_id, answer_url, answer_text, word_count
• upvotes / upvotes_text, view_count / view_count_text
• comments_count / comments_count_text, shares_count / shares_count_text
• created_at, answer_created_at, answer_timestamp, answer_updated_at
• author_name, author_url, author_id, author_image_url, author_credentials, author_is_verified, author_is_anonymous
• is_ai_answer, ai_label_text, answer_visibility, answer_is_collapsed
• answer_is_related, source_question_url, source_question_title
• question_id, question_title, question_url
• media_urls, image_urls, video_urls, outbound_links (optional)
• answer_html (optional), discovery_source, source_url, access_state

Output per author

• author_id, name, profile_url, avatar_url, credential, is_verified
• follower_count, following_count, answer_count, question_count, post_count, content_views
• joined_at, location, employment, education
• topics[], bio, website_url, social_links[]
• evidence_urls[], evidence_question_urls[], evidence_answer_urls[], evidence_answer_previews[]

A direct profile URL in urls mode emits a profile row with the same profile fields. In authors mode it emits an author row.

Output per topic

• topic_id, name, url, follower_count
• discovery_source, source_url, access_state

Output per status / manifest

Status rows explain a page that could not be read: access_state is one of not_found, removed, login_required, cloudflare_challenge, rate_limited, unavailable, unsupported_url, or discovery_unavailable. A run-manifest row is written at the end of every run that starts.

Input

Field	Type	Default	Description
mode	string	urls	urls, search, or authors
directUrls	array	– (prefilled with one Python question)	Question, answer, profile, or topic URLs
searchQueries	array	–	Keywords or question titles for best-effort discovery (mode=search)
usernames	array	–	Profile slugs converted to profile URLs
resultTypes	array	["question","answer"]	question, answer, profile, author, topic
includeAnswers	boolean	true	Emit answers from question pages
maxAnswersPerQuestion	integer	20	Answer cap per question (0–500)
includeRelatedQuestions	boolean	false	Follow related-question links
maxRelatedQuestions	integer	15	Cap per question; 0 keeps all observed
includeAuthorProfiles	boolean	true	Enrich discovered authors with profile stats
maxAuthors	integer	100	Author-row cap
minAnswerUpvotes	integer	0	Drop answers below this explicit upvote count
filterAiAnswers	string	include	include, exclude, or only
answerSortOrder	string	observed	observed, upvotes_desc, or newest
onlyUnanswered	boolean	false	Keep only questions with an explicit zero answer count
includeMedia	boolean	false	Retain answer image/video links
includeOutboundLinks	boolean	false	Retain external links in answers
includeAnswerHtml	boolean	true	Retain a simple HTML rendering of answers
includeProfileSensitiveFields	boolean	true	Include location, employment, education, join date
includeSocialLinks	boolean	true	Include profile website/social links
maxItems	integer	200	Total real-row cap
maxItemsPerSource	integer	0	Per-URL/keyword cap (0 = no cap)
includeFieldEvidence	boolean	true	Add field_sources and fields_present

The saved prefill is one question with maxAnswersPerQuestion: 2 and maxItems: 3, so the first run returns three real rows.

Examples

A question with many answers

{
  "mode": "urls",
  "directUrls": ["https://www.quora.com/What-is-Python-primarily-used-for"],
  "resultTypes": ["question", "answer"],
  "maxAnswersPerQuestion": 50
}

A profile and its stats

{
  "mode": "authors",
  "directUrls": ["https://www.quora.com/profile/Adam-DAngelo"],
  "resultTypes": ["author"]
}

Expand related questions from one seed

{
  "mode": "search",
  "directUrls": ["https://www.quora.com/What-is-machine-learning"],
  "includeRelatedQuestions": true,
  "maxRelatedQuestions": 10,
  "maxItems": 100
}

Best-effort keyword discovery

{
  "mode": "search",
  "searchQueries": ["remote work productivity"],
  "includeRelatedQuestions": true,
  "maxItems": 50
}

Use cases

• Market and product research — collect complete answer threads with engagement metrics.
• Expert sourcing — turn answer authors into a profile list with follower and answer counts.
• Content gap analysis — find questions with zero answers in a niche.
• AI-answer auditing — separate explicitly AI-labelled answers from human answers.
• Knowledge-base building — export clean question/answer pairs with source URLs and dates.

FAQ

Does this actor need cookies or a login?

No. It reads public question, answer, profile, and topic pages without an account. There is no cookie or proxy field to configure.

Why can't I search Quora by keyword reliably?

Quora's own search page is gated for anonymous visitors, and the actor never calls Quora's /search?q= endpoint. It uses a public web index for keyword discovery and expands related questions from any question it reaches. When no indexed matches exist, it writes a discovery_unavailable status row and does not invent results.

How many answers can I get?

All answers that Quora exposes through its answer feed, up to maxAnswersPerQuestion. The actor follows the answer cursor until your cap or the end of the feed. Popular questions can contain thousands of answers; set a cap for the first run.

Why is an answer marked as related?

Quora sometimes shows answers written for a merged or equivalent question. Those rows carry answer_is_related: true and source_question_url / source_question_title, so the answer is not misattributed to the page you opened.

Why are some fields missing on some rows?

Quora only publishes a credential, location, employment, education, or bio for some users. Missing values are omitted, never guessed. Question creation dates and question view counts are not published on the public question page and are not emitted.

Can I get comment text?

No. Quora publishes the comment count with each answer, but not the comment bodies on the public page. The actor returns the count only.

What does "best-effort" mean for keyword discovery?

It means the actor tries a public index first and tells you when that index has no matches. It never re-labels related questions or unrelated pages as keyword search results.

What happens if Quora challenges the page?

The actor first tries its fast HTTP path, then datacenter and residential HTTP. If all three are challenged, it renders that page in a real browser through its residential connection and parses the same data. If even that is blocked, the page is reported as a cloudflare_challenge status row. The browser fallback is slower, uses more memory, and incurs residential data transfer for the pages that need it; it only runs after direct, datacenter, and residential HTTP are challenged.

Is this affiliated with Quora?

No. This is an independent, third-party actor that reads public Quora pages. It is not affiliated with or endorsed by Quora.

Limitations

• Native Quora keyword search is unavailable anonymously; keyword discovery depends on a public web index.
• Comment bodies, question creation dates, question view counts, topic question feeds, and full profile answer lists are not available from the public page.
• Topic pages expose the topic name and follower count; descriptions and question counts are omitted when Quora does not publish them.
• Very large answer threads are bounded by maxAnswersPerQuestion and maxItems.
• A page that is removed, private, or behind a login wall is reported as a status row, not fabricated data.
• Challenge recovery uses a real browser as a last resort, so it is slower, uses more memory, and incurs residential data transfer for the pages that need it.