Python Twitter scraper: scrape tweets with an API
Scrape Twitter with Python by calling an API instead of driving a browser. The script below uses requests to pull a user's recent tweets or run a search and print them. There is no browser to automate, no login and no HTML to parse. It reads public X tweets only.

The job
A scraper that opens X in a headless browser breaks when the page changes and gets blocked when it logs in. Calling an API avoids both problems. You send a GET request and read JSON, so the script stays short and keeps working.
This page covers getting tweets out of X for analysis, a dataset or a monitor. The X API in Python guide goes deeper on call patterns and credit use. The script here is the quick start: copy it, set one environment variable and run it.
You can use it to scrape a user's tweets, search a keyword or hashtag, or build a small dataset in a loop. It returns public tweets only, one page per call, and a cursor takes you to the next page.
How it works
Install requests
Run
pip install requests. You don't need anything else.Set your key
Export
CHATTERAPI_KEYin your shell. A test key returns recorded sample data, so you can run the script before you pay for anything.Run the script
Save the code as
scrape.pyand runpython scrape.py levelsiofor a user's tweets, or edit it to call search.Page and save
Follow
page.nextCursorfor more rows, then write the list to a JSON or CSV file for your analysis.
Endpoints that do it
| Endpoint | What it's for | Credits |
|---|---|---|
| List user tweetsGET /v1/x/users/{handle}/tweets | A user's recent tweets, newest first. | 2 credits |
| Search tweetsGET /v1/x/tweets/search | Tweets by keyword or hashtag, with X search operators. | 2 credits |
| Get a tweetGET /v1/x/tweets/{id} | One tweet in full, by id or link. | 2 credits |
Free tools for the same job
Try the job in a browser first. These tools call the same endpoints, and you don't need a key or any code.
What the script prints
The script above calls the timeline endpoint. This is a recorded response from the same call.
import os
import sys
import requests
BASE = "https://api.chatterapi.com/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['CHATTERAPI_KEY']}"}
def user_tweets(handle, max_pages=3):
"""Collect recent tweets for a handle, following the cursor."""
rows, cursor = [], None
for _ in range(max_pages):
params = {"cursor": cursor} if cursor else {}
r = requests.get(f"{BASE}/x/users/{handle}/tweets", headers=HEADERS, params=params, timeout=60)
r.raise_for_status()
body = r.json()
rows += body["data"]
cursor = body["page"]["nextCursor"]
if not body["page"]["hasMore"] or not cursor:
break
return rows
if __name__ == "__main__":
for t in user_tweets(sys.argv[1] if len(sys.argv) > 1 else "levelsio"):
print(t["createdAt"], t["likes"], t["text"][:80].replace("\n", " "))
curl "https://api.chatterapi.com/v1/x/users/levelsio/tweets" \
-H "Authorization: Bearer $CHATTERAPI_KEY"{
"data": [
{
"id": "2110066474513793361",
"text": "✨ 200,000+ users on my computer nostalgia project @pieter\n\nSometimes it takes a few years for things to take off, in this case about 2 years!\n\nKeep creating 💖 https://t.co/Na0LXnwAjW",
"url": "https://x.com/levelsio/status/2110066474513793361",
"createdAt": "2026-10-13T17:53:45.000Z",
"lang": "en",
"likes": 129,
"retweets": 4,
"replies": 37,
"quotes": 0,
"views": 15475,
"bookmarks": 9,
"isReply": false,
"inReplyToId": null,
"conversationId": "2110066474513793361",
"isRetweet": false,
"quotedTweetId": "2108957488410161442",
"type": "quote",
"isQuote": true,
"isPinned": false,
"source": null,
"hasVideo": false,
"urls": [],
"mentions": [
"pieter"
],
"author": {
"id": "1577241403",
"handle": "levelsio",
"name": "@levelsio",
"bio": "💸http://levels.vc\n📸http://PhotoAI.com $86K/m\n🎮http://VibeJam.com $44K/m\n🏡http://InteriorAI.com $23K/m\n👙@X $13K/m\n🌍http://Nomads.com $7K/m\n🏩http://Hotelist.com $5K/m",
"followers": 974707,
"following": 3276,
"tweets": 181450,
"verified": false,
"blueVerified": true,
"canDm": false,
"protected": false,
"location": "📕My book $5K/m 👉",
"url": "https://readMAKE.com",
"avatarUrl": "https://pbs.twimg.com/profile_images/2077111020305162240/PwddgOau_400x400.jpg",
"createdAt": "2013-07-08T09:31:59.000Z",
"profileUrl": "https://x.com/levelsio",
"bannerUrl": "https://pbs.twimg.com/profile_banners/1577241403/1665964220",
"likesGiven": 185852,
"mediaCount": 19829,
"pinnedTweetIds": [
"2109949585737052316"
]
},
"media": [
{
"type": "photo",
"url": "https://pbs.twimg.com/media/HUh2NAYWQAAP7ZK.jpg",
"videoUrl": null
}
],
"quotedTweet": {
"id": "2108957488410161442",
"text": "💾 Two years ago I started working on these crazy nostalgia projects, kind of randomly, because I was bored and wanted to do something different than just build another SaaS or AI app/website\n\nI wanted to try to run Wind…",
"url": "https://x.com/levelsio/status/2108957488410161442",
"createdAt": null,
"lang": null,
"likes": null,
"retweets": null,
"replies": null,
"quotes": null,
"views": null,
"bookmarks": null,
"isReply": false,
"inReplyToId": null,
"conversationId": null,
"isRetweet": false,
"quotedTweetId": null,
"type": "tweet",
"isQuote": false,
"isPinned": false,
"source": null,
"hasVideo": false,
"urls": [],
"mentions": [],
"author": {
"id": "",
"handle": "levelsio",
"name": "@levelsio",
"bio": null,
"followers": null,
"following": null,
"tweets": null,
"verified": null,
"blueVerified": null,
"canDm": null,
"protected": null,
"location": null,
"url": null,
"avatarUrl": null,
"createdAt": null,
"profileUrl": "https://x.com/levelsio",
"bannerUrl": null,
"likesGiven": null,
"mediaCount": null,
"pinnedTweetIds": null
},
"media": []
},
"retweetedTweet": null
},
{
"id": "2110030627336540370",
"text": "This is also interesting to think about\n\nRight now it still takes, let's say, 30 minutes to a few hours to build something usable with AI and then you still polish for days/weeks/months\n\nIf we assume this will keep getti…",
"url": "https://x.com/levelsio/status/2110030627336540370",
"createdAt": "2026-10-13T15:31:18.000Z",
"lang": "en",
"likes": 324,
"retweets": 11,
"replies": 112,
"quotes": 10,
"views": 38339,
"bookmarks": 118,
"isReply": false,
"inReplyToId": null,
"conversationId": "2110030627336540370",
"isRetweet": false,
"quotedTweetId": null,
"type": "quote",
"isQuote": true,
"isPinned": false,
"source": null,
"hasVideo": false,
"urls": [],
"mentions": [],
"author": {
"id": "1577241403",
"handle": "levelsio",
"name": "@levelsio",
"bio": "💸http://levels.vc\n📸http://PhotoAI.com $86K/m\n🎮http://VibeJam.com $44K/m\n🏡http://InteriorAI.com $23K/m\n👙@X $13K/m\n🌍http://Nomads.com $7K/m\n🏩http://Hotelist.com $5K/m",
"followers": 974707,
"following": 3276,
"tweets": 181450,
"verified": false,
"blueVerified": true,
"canDm": false,
"protected": false,
"location": "📕My book $5K/m 👉",
"url": "https://readMAKE.com",
"avatarUrl": "https://pbs.twimg.com/profile_images/2077111020305162240/PwddgOau_400x400.jpg",
"createdAt": "2013-07-08T09:31:59.000Z",
"profileUrl": "https://x.com/levelsio",
"bannerUrl": "https://pbs.twimg.com/profile_banners/1577241403/1665964220",
"likesGiven": 185852,
"mediaCount": 19829,
"pinnedTweetIds": [
"2109949585737052316"
]
},
"media": [],
"quotedTweet": {
"id": "",
"text": "",
"url": "https://x.com/i/status/",
"createdAt": null,
"lang": null,
"likes": null,
"retweets": null,
"replies": null,
"quotes": null,
"views": null,
"bookmarks": null,
"isReply": false,
"inReplyToId": null,
"conversationId": null,
"isRetweet": false,
"quotedTweetId": null,
"type": "tweet",
"isQuote": false,
"isPinned": false,
"source": null,
"hasVideo": false,
"urls": [],
"mentions": [],
"author": {
"id": "",
"handle": "",
"name": "",
"bio": null,
"followers": null,
"following": null,
"tweets": null,
"verified": null,
"blueVerified": null,
"canDm": null,
"protected": null,
"location": null,
"url": null,
"avatarUrl": null,
"createdAt": null,
"profileUrl": "https://x.com/",
"bannerUrl": null,
"likesGiven": null,
"mediaCount": null,
"pinnedTweetIds": null
},
"media": []
},
"retweetedTweet": null
}
],
"page": {
"nextCursor": "AYCt…",
"hasMore": true
}
}Recorded from a live call on . Long text is shortened and the cursor is cut. A live key is charged 2 credits for this call. A test key returns this same sample for 0 credits.
What it won't do
- Public tweets only. Protected accounts and deleted tweets are not returned.
- Each call returns one page of about 20 tweets. The script caps pages with
max_pagesso a loop cannot spend more than you intend. - Keep the key in an environment variable and out of source control.
- It reads tweets. It does not scrape Instagram, TikTok or other sites.
Questions
How do I scrape Twitter with Python without the official API?
requests. You skip the developer app and OAuth setup and read the JSON that comes back.Do I need Selenium or a login to scrape tweets?
Can I scrape tweets from a hashtag or keyword?
/x/tweets/search and pass your query as q, for example q=%23python.How do I save the tweets?
json.dump, or load it into pandas with pandas.DataFrame(rows) for analysis.