Twitter (X) Tweets & Profiles Scraper

Scrape posts and full profile data from X (Twitter) accounts — real post text, timestamps, likes, retweets, replies, quotes, bookmarks, views, media, links, and profile follower counts. No X API key required.

The Actor talks to X's own GraphQL API using an anonymous guest token, so it is fast (a 40-post scrape typically finishes in well under a minute) and does not need a headless browser.

---

What it extracts

Per post (itemType: "tweet")

Field	Description
id, url	Post ID and canonical x.com URL
text	Full post text (long posts are un-truncated from note_tweet)
timestamp, createdAt	X's raw timestamp and the same moment as UTC ISO-8601
likes, retweets, replies, quotes, bookmarks, views	Live engagement counts
isPinned, isRetweet, isReply, isQuote	Post type flags
inReplyToTweetId, inReplyToUsername, conversationId	Threading info
quotedTweetId, retweetedTweetId	IDs of the quoted / retweeted post
links, media	Expanded URLs, and photos/videos/GIFs with direct media URLs
lang	Language X detected
username, fullname, verified, isBlueVerified, isVerified, verifiedType	Author identity
user	Full nested author profile (see below) — toggle with Include author profile

Engagement figures on a retweet are those of the original post: X sends literal zeros on the retweet wrapper itself, which would otherwise look like a post with no engagement.

Per profile (itemType: "profile")

userId, username, fullname, profileUrl, description, location, website, joinDate / joinDateIso, avatar, banner, isProtected, pinnedTweetIds, totalFollowers, totalFollowing, totalTweets, totalMediaCount, totalLikes, and three separate verification fields.

Verification is three different things and this Actor keeps them apart:

• isBlueVerified — the account pays for X Premium (the checkmark most accounts have today).
• isVerified — the old legacy verification flag, which is false for nearly every account now.
• verifiedType — Government, Business, … for organisation checkmarks, otherwise null.
• verified — convenience flag: true when the account displays any checkmark.

A count X does not return comes back as null, never as 0.

---

Input

Option	What it does
Targets (startUrls)	Profile URLs (https://x.com/NASA), usernames (NASA, @NASA), numeric user IDs (11348282) or keywords. One per line.
Maximum posts per target (maxTweets)	Hard cap on posts collected per target, 1–1000.
What to output (resultType)	tweets, profiles, or both.
Sort order (sortOrder)	newest, oldest, mostLikes, mostRetweets. Non-chronological orders gather a pool of up to 200 posts before sorting.
Include replies (withReplies)	Keep or drop reply posts.
Include retweets (includeRetweets)	Keep or drop reposts.
Include author profile (includeUserInfo)	Attach the nested user object to every post.
Maximum run time (maxRunTimeSecs)	Wall clock. Paging stops when it is reached and everything collected so far is kept. Automatically trimmed to finish inside the platform run timeout.
authToken / csrfToken	Optional login cookies — see the limits below.
Proxy configuration	Optional. Falls back to RESIDENTIAL, then to a direct connection, if the chosen group is unavailable.

---

What X allows, honestly

Anonymous access to X is not unlimited, and this Actor states plainly what is and is not reachable rather than quietly returning empty results:

• Profile data (UserByScreenName) — works anonymously, including live follower counts.
• Account timelines (UserTweets) — works anonymously, with cursor pagination.
• Keyword / hashtag search (SearchTimeline) — X answers anonymous callers with HTTP 404. Keyword targets are skipped with a warning unless you supply the optional authToken + csrfToken cookies from a logged-in x.com session.
• Replies to other accounts (UserTweetsAndReplies) — also login-gated. Anonymously you get the account's own timeline, which includes its self-thread replies.
• For very large accounts X serves logged-out clients a ranked "top posts" page with no pagination cursor, so those accounts return roughly 100 posts however high maxTweets is.

If a run collects nothing it fails with a status message explaining why — it never reports a green run over an empty dataset.

---

Example output

{
  "itemType": "tweet",
  "id": "2084359126121853051",
  "url": "https://x.com/NASA/status/2084359126121853051",
  "username": "@NASA",
  "fullname": "NASA",
  "verified": true,
  "isBlueVerified": true,
  "isVerified": false,
  "verifiedType": "Government",
  "createdAt": "2026-08-03T19:21:55Z",
  "text": "Is your name part of our legacy of exploration and discovery?",
  "likes": 1028,
  "retweets": 209,
  "replies": 133,
  "quotes": 16,
  "bookmarks": 67,
  "views": 491875,
  "isRetweet": false,
  "isReply": false,
  "lang": "en",
  "user": {
    "username": "NASA",
    "userId": "11348282",
    "totalFollowers": 92238355,
    "totalFollowing": 119,
    "totalTweets": 74288,
    "joinDateIso": "2007-12-19T20:20:32Z",
    "location": "Pale Blue Dot",
    "website": "http://www.nasa.gov/"
  }
}

---

Legal

Scrape only public data and use it in line with X's terms and applicable law (including GDPR where personal data is involved).