What is X (Twitter) Scraper?
X (Twitter) Scraper runs an X search, profile, list or post link for you and returns the matching posts, one record per post. Write the search the way you would on x.com, with operators such as from:, since: or min_faves:, or fill in filters for author, mentions, language, dates and minimum engagement.
Each post carries its full text, engagement counts, media links and the author's profile, so one run tells you both what was said and who said it.
You ask
Get SpaceX's 20 latest posts on X with views, likes, reposts and replies.
It runs
From your AI assistant over MCP or from your code over REST
You get
- Post
- Engagement
- Conversation
- Media
- Author
- Follow a launch as it happensPull posts from your brand, competitors or partners during a launch day and see which ones travel.
- Research a topic or productCollect posts that mention a product, a ticker or a phrase, in one language and one date range.
- Find the voices in a nicheFilter by minimum likes or verified authors to see whose posts people actually read and share.
- Keep a record of an accountSave a profile's recent posts with dates, counts and media for a report or an internal archive.
- Brief your agentLet Claude or ChatGPT collect the posts, then cluster the questions people ask or draft a reply plan.
What data does the Twitter scraper extract?
Each post comes back as one record with its author attached. These are the fields our test run returned.
Post
- Post ID and URL
- Full text
- Language
- Date and time
- Original, reply, quote or repost
- Pinned flag
- Sensitive media flag
Engagement
- Views
- Likes
- Reposts
- Quotes
- Replies
- Bookmarks
Conversation
- Conversation ID
- Replied-to post and account
- Quoted post with its author
- Reply restrictions flag
Media
- Image and video links
- Video streams in several qualities
- Media sizes
- Links in the text
Author
- Name and handle
- Bio, location and website
- Followers and following
- Verification type
- Post and media counts
- Account creation date
Fields appear when a post has them: a text-only post has no media, and only replies name a replied-to account. For the replies under one post, use X (Twitter) Post Replies.
Sample output from SpaceX
The table at the top of this page is a real run on 16 November 2026: the search from:SpaceX in Latest order, 20 posts requested and 20 returned. They cover 14 to 16 November, during Starship Flight 15 and the return to Starbase. Below the chart is the most viewed post in full.
One result in full
Post 6 in full
- Link
- x.com/SpaceX/status/2121715216597827756
- Text
- Mechazilla catches Starship! Congratulations to the entire SpaceX team on this incredible achievement!
- Posted
- 14 Nov 2026, 21:21 UTC
- Type and language
- Original post, English
- Media
- 1 video, with stream and MP4 links
- Views
- 16,572,449
- Likes
- 139,863
- Reposts and quotes
- 27,188 and 9,035
- Replies
- 6,767
- Bookmarks
- 13,768
- Author
- SpaceX (@SpaceX), 41,970,067 followers, business verification
- Account since
- April 2009
What these 20 posts tell you
One post, the catch announcement, drew 16.6 million of the 41.4 million views across all 20, or 40%. Report the median next to the total, here 1.2 million views.
The three posts under two hours old show likes at 3.2% to 3.7% of views. The other 17 range from 0.2% to 1.6%. Compare posts of similar age.
17of 20
17 of 20 posts carry media and 11 carry video. Three are quote posts of SpaceX's own livestream, one is a reply in its own thread, and none are reposts.
Bookmarks beat replies on all 20 posts, so bookmarks are worth a column of their own when you rank content.
Scrape X posts in four steps
Connect AI/ML API
Click Try it now, choose Claude, ChatGPT, Cursor or another MCP client and add the server. You sign in once.
Describe the posts
Name an account, a search phrase, a list or a post link, plus the period and how many posts you want.
Get the post table
Your assistant runs the search and lists each post with views, likes, reposts, replies and time.
Analyze in the same chat
Ask it to chart engagement by hour, find the most quoted post or pull the replies to the top one.
Prompts to copy
Use the AI/ML API MCP X (Twitter) Scraper tool. Get the 20 latest posts from @SpaceX and show date, views, likes, reposts and replies in a table.
Use the AI/ML API MCP X (Twitter) Scraper tool. Find English posts from the last 7 days that mention “heat pump” and have at least 50 likes, sorted by Top.
Use the AI/ML API MCP X (Twitter) Scraper tool. Collect posts mentioning @Ryanair this week and group the complaints by topic, with two examples each.
Use the AI/ML API MCP X (Twitter) Scraper tool. Compare the last 30 posts of @Nike and @adidas: median views, median likes and how many carry video.
Search settings for X posts
Ask in plain words and your assistant fills these in. In code you pass them yourself.
| Setting | What it controls | Tip |
|---|---|---|
| Search terms | Words, hashtags or advanced operators such as from:, since: and min_faves: | One idea per search term keeps results clean |
| Start URLs | Links to posts, profiles, searches or lists | Paste a search you already tuned on x.com |
| Author, mentioning, in reply to | Posts from, mentioning or replying to a handle | Pair with dates for a clean window |
| Start and end date | Only posts sent after and before the dates you set | Use for launch days and events |
| Sort | Top, Latest, or both at once | If Latest returns few posts, try Top |
| Engagement and media filters | Minimum likes, reposts or replies; only images, videos or quotes; only verified authors | Raise the minimum to skip low-signal posts |
| Language and location | Post language, posts near a point or within a radius, or tagged with a place | Language uses two-letter codes such as en or de |
| Maximum posts | Cap on the number of posts returned | Start with 20 to check relevance |
Latest + Top runs both orders at once to collect more posts, so the same post can appear twice. Remove duplicates by post ID before you count anything.
What teams do with X post data
Launch tracking
Pull your own and competitors' posts during a launch window and see which messages earned reposts and quotes.
DateViewsRepostsQuotes
Brand mentions
Collect posts that mention your handle or product name and send complaints to the support queue.
TextAuthorDate
Competitor content review
Compare what competitors post, how often, and which formats carry their views.
ViewsLikesMedia
Influencer discovery
Search a topic with a minimum like count and list the authors who keep showing up, with their follower counts.
AuthorFollowersLikes
Event timelines
Pull posts for a hashtag and a date window to reconstruct what was said and when.
DateTextQuotes
Research datasets
Collect posts in one language and period for analysis, labeling or testing a classifier.
TextLanguageDate
Twitter scraper API for your code
In your code
REST · your AI/ML API key, JSON records back
The same scraper runs over REST with your AI/ML API key. Send search terms or links with a cap and filters, and you get JSON posts with the fields above, ready for a dashboard or an alerting job.
Use MCP when a person explores a topic in a chat and REST when a job runs the same searches every hour. Both run on one AI/ML API balance, without an X developer account. Start with the Toolbox REST quickstart.
Other ways to collect X posts
The method that fits depends on how many posts you need and whether the work happens in code or in a chat.
| Copy posts by hand | X API | Your own scraper | ||
|---|---|---|---|---|
| Setup | Connect once with your AI/ML API account | None | Developer account and a paid plan with usage caps | Accounts, sessions, proxies and frequent fixes |
| Inside an AI chat | Yes. Over MCP | No. No | Only if you build it | Only if you build it |
| Best for | Searches, profiles and lists from chat or code | A few posts for a slide | Products that must use X's official API | Teams with engineers to maintain it |
Pricing
Pay as you goBilled per run from your AI/ML API balance
- No subscription
- No seat fee
- One balance for every tool
Each search is billed per run from your AI/ML API balance, with no monthly plan. The price follows the number of posts returned. Our 20-post test run cost less than one cent. Check the current price in your dashboard before runs of thousands of posts.
FAQ
Is it legal to scrape Twitter?
The tool reads public posts. Collecting them for research and monitoring is common, but X's terms apply to the service, and privacy laws such as GDPR apply when you store personal data about the people who post. Ask a lawyer about your specific case.
How many posts can I get from one search?
Set a maximum and the scraper stops there. If Latest returns fewer posts than you expected, switch to Top, or to Latest + Top and remove duplicates afterwards.
Can I get the replies to a post?
Each post includes its reply count, and you can search replies with the in-reply-to filter or a conversation ID. For the reply thread under one post, use X (Twitter) Post Replies.
Does it return profile data?
Every post includes its author's profile: followers, following, bio, location, website, verification type and account creation date. For one account's profile on its own, use X (Twitter) Profile Details.
How is this different from the X API?
The X API is X's own developer product and needs a developer account and a paid plan with usage caps. X (Twitter) Scraper returns public posts on your AI/ML API balance and works inside AI assistants over MCP.
Can I use it from Claude, ChatGPT or Cursor?
Yes. Add the AI/ML API MCP server once with the MCP quickstart and ask in plain English. Other MCP clients work the same way.
Can I export posts to CSV or Google Sheets?
Ask your assistant to return the table as CSV, or write the JSON from the REST response to a file or a sheet in your code.
How fresh are views and likes?
Each run reads X at that moment, and counts keep moving after a post goes out. In our sample the three newest posts showed likes at over 3% of views, against about 1% for older ones, so compare posts of similar age.
Run your first X search
Connect once, name an account or a topic and get the posts as a table in your chat.