WEIR

Find Tweets That Link to a URL (BackTweets Alternative)

Find tweets that link to a URL with url:yourdomain.com on X, tested on nasa.gov, plus a 15-line Python export and alerts when a new post links to you.


Published · Updated · 7 min read

Cover image: Who linked to my site?

To find tweets that link to a URL, search url:yourdomain.com on X (Twitter), then open the Latest tab.

This is the BackTweets alternative for anyone who asks "who shared my link on Twitter?" and wants author, date and likes. To export or watch it, run the same search through an API. We run Weir, the API in the examples.

Paste this into X's search box and swap in your domain:

url:yourdomain.com -filter:replies
  1. Search url: on XType url:yourdomain.com and open Latest.
  2. Narrow itAdd a path, -filter:replies or a date range.
  3. Ask your AI appPaste the prompt below, with Weir connected.
  4. Export or watch itRun the 15-line script, or save the search as a monitor.

What happened to BackTweets?

BackTweets was a free site that showed every tweet linking to a URL. BackType launched it on 6 March 2009, and Twitter bought BackType on 5 July 2011, according to Wikipedia and TechCrunch's report.

The service then closed. Sorezki Ltd. bought the domain at auction in March 2012 and relaunched it.

On 16 November 2026, backtweets.com loads a page that says "Paste a link. See where it travels." It does not say which networks it covers, and we could not confirm it still lists X posts.

The job is the same today, and X's own search does it.

The operator is url:. X's developer docs describe it as a "tokenized match" on a post's url or expanded_url fields. So it reads the real address behind a short link.

Our operator test of 27 October 2026 found url:github.com returned 10 matching posts out of 10. On 16 November 2026, every post for url:nasa.gov carried a nasa.gov link in its entities.urls.

Here are the forms we tested through Weir on 16 November 2026:

You type What we saw
url:nasa.gov Posts with nasa.gov links, also with a subdomain
url:"nasa.gov/humans-in-space/astronauts" Only posts with that path, 1 post since 1 Nov
url:mercedes-benz.com 5 of 5 posts had a mercedes-benz.com link
url:mercedes_benz.com 0 posts

The hyphen row matters.

A reference we had read says to use underscores, not hyphens, inside url: values. For a real domain with a hyphen, that is wrong: the underscore version found nothing, and the plain hyphen worked.

A short link is matched through its expanded address, so you never type t.co. The match is also case-insensitive: a post with "Nasa.gov" showed up.

Type the operator into the search box on x.com, then choose the Latest tab so you see posts in time order. Add these to tighten it:

  • -filter:replies drops replies, so you see the posts that start a conversation.
  • filter:links keeps only posts that contain a link.
  • since:2026-11-01 until:2026-11-15 sets a date range. until: includes its own day.

Our free X advanced search builder writes these from a form. Operator test results and the operators guide cover the rest.

We did not test what x.com shows when you are signed out.

If your AI app has Weir connected over MCP, you can skip the syntax. Paste this and change the domain:

Find posts from the last 7 days that link to example.com. Group them by
account, show the most-liked first, and give each post's date and link.

We ran that search for nasa.gov on 16 November 2026, with a start of 9 November. It returned 15 posts for 16 credits, from 9 accounts. They had 616 likes in total, and the most-liked post had 287.

Only one of the 15 came from a NASA account: @NASApeople, on 12 November, with 4 likes. Most of the rest were small accounts sharing lists or birthday posts, and one account posted 4 of them.

A few posts carry most of the likes. The free tweet finder runs a search like this on your own Weir account, if you would rather not use a chat.

Export every post that links to a URL with Python

This script uses only the standard library. It pages through url: search between two dates, drops duplicate ids, and stops on a null cursor or after 3 pages. It writes posts.csv with the post link, author, likes and date.

import csv, json, os, sys, urllib.request
domain, since, until = sys.argv[1:4]
body = {"query": f"url:{domain} -filter:replies", "since": since, "until": until, "limit": 20}
seen, rows = set(), []
for _ in range(3):  # page cap
    req = urllib.request.Request("https://api.weirapp.io/v1/x/search", json.dumps(body).encode(),
        {"Authorization": "Bearer " + os.environ["WEIR_API_KEY"], "Content-Type": "application/json"})
    page = json.load(urllib.request.urlopen(req, timeout=60))
    for p in page["data"]:
        if p["id"] not in seen:
            seen.add(p["id"]); rows.append([p["url"], p["author"]["username"], p["likeCount"], p["createdAt"][:10]])
    if not page["meta"]["nextCursor"]: break
    body["cursor"] = page["meta"]["nextCursor"]
csv.writer(open("posts.csv", "w", newline="")).writerows([["url", "author", "likes", "date"]] + rows)
print(len(rows), "posts")

Run it with python url_posts.py nasa.gov 2026-11-13T00:00:00Z 2026-11-16T00:00:00Z. We ran it on 16 November 2026 and it wrote 4 posts. The file starts like this, with a personal account blanked out:

url,author,likes,date
https://x.com/<account>/status/<id>,<account>,0,2026-11-15

That run charged 5 credits for 4 posts, then 1 for an empty page. The first page said more could follow, which is why the loop stops only on a null cursor.

Four posts in three days is small. X search misses posts, so treat the CSV as a lower bound.

A search page of 20 posts costs 21 credits: 1 for the call and 1 per post.

21credits for a page of 20 posts
5,000free credits every month, no card
$5buys 25,000 credits that never expire

The free credits cover about 238 full pages a month. Price a bigger job with the X API cost calculator.

Save the search as a Weir monitor. A monitor runs your query on a schedule, keeps only results it has not shown you, and sends them to a webhook.

curl --request POST --url https://api.weirapp.io/v1/monitors \
  --header "Authorization: Bearer $WEIR_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{"name":"Links to my site","target":{"kind":"search","query":"url:yourdomain.com -filter:replies","sources":["x"]},"schedule":"daily"}'

Create a webhook with POST /v1/webhooks, then point the monitor at it with webhookId. A Slack, Discord or Telegram incoming-webhook URL works too. The monitors and webhooks docs have the steps, and our Twitter webhooks page explains the idea.

As of 16 November 2026, the docs say a search monitor reads one page of about 20 posts per run. hourly is for paid accounts, and the free tier keeps one monitor.

For mentions without a link, see how to see mentions on X. For many keywords, see the social listening API.

Caveats and when not to use Weir

  • X search skips posts. X says search may not show every post, so read why X search is missing results. A match list is a floor, not a census.
  • Protected accounts never appear. Their posts are visible only to approved followers.
  • Deleted posts disappear. A post removed since a past export will not be there on the next run.
  • You need only a quick look. X's own search box costs nothing, so use it.
  • You need a contract with X, or to post. Weir reads public data only and cannot post.

Weir is not licensed by X, so access can change. Check X's Terms of Service for your use. This is not legal advice.

FAQ

Is there a BackTweets alternative that still works?

BackTweets itself loads a page today, but we could not confirm it still lists X posts. The alternative is X's own url: search, or the same query through Weir for a CSV, an AI chat or alerts. We run Weir.

Do hyphens in a domain work in url: searches?

Yes. On 16 November 2026, url:mercedes-benz.com returned 5 posts out of 5 with that domain in their links. The underscore version, url:mercedes_benz.com, returned none.

How do I search Twitter for links to a specific article, not the whole site?

Put the path in quotes: url:"nasa.gov/humans-in-space/astronauts". When we tried that path on 16 November 2026, it returned the one post whose link started with it.

Sources

Checked 16 November 2026: