Threads Posts, Search & Replies Scraper

Collect Threads posts three ways — the latest posts of any public profiles, the top posts for any keyword, and any post you link with the replies under it — with exact likes, replies, reposts, quotes and shares, text, photos and videos.

No Threads or Instagram login, no cookies, no API key.

What it's for

• Brand and topic monitoring — run daily on your keywords and competitors' accounts and catch new posts while they are fresh.
• Campaign reporting — paste the links of sponsored posts and get exact engagement for each, ready for the client report.
• Influencer research — see what a creator posts and how their audience responds.
• Audience research — read the replies under a brand's or a creator's post: questions, complaints, praise.
• Content research — what is working on Threads for a topic right now.

What you get for each post

Engagement	Exact likes, replies, reposts, quotes and shares (not rounded)
Content	Text, media type (text, photo, video, carousel), photo and video URLs, links, @mentions, hashtags, link preview
Context	Quoted or reposted post, whether it is a reply and to whom, paid-partnership flag, edited flag, language
Account	Username, display name, verified badge, user ID
Post	Post ID, short code, link, date and time; where it came from (profile, search or post) and whether it is a post or a reply

Real rows from a test run:

Account	Source	Text	Likes	Replies	Reposts
nasa	search "nasa"	Spotted: views of a "space jellyfish"…	39,525	—	—
zuck	profile	Introducing Muse Wonder, our new family of frontier models…	1,928	306	137
cristi.baluta	reply to zuck	Can you make that stupid moose to remove…	16	1	0

Three ways to collect

Input	What comes back	Measured
usernames	The account's latest threads, as Threads shows them to visitors	4–10 posts per account
searchQueries	The top posts Threads shows for the keyword, from any account	22–30 posts per keyword
postUrls	The post itself plus the replies shown under it, each with its own likes	up to ~15 replies per post

Use any combination in one run. A post found twice (for example in a profile and a search) is saved once and charged once.

Input

{
  "usernames": ["zuck", "@nasa", "https://www.threads.com/@netflix"],
  "searchQueries": ["pickleball", "world cup"],
  "postUrls": ["https://www.threads.com/@zuck/post/DeZtp-TkQHh"],
  "includeReplies": true
}

Field	Default	Notes
usernames	—	Usernames or profile links.
searchQueries	—	One keyword per line.
postUrls	—	threads.com or threads.net post links.
includeReplies	true	Save the replies under linked posts.
includeOwnThreadReplies	true	On profiles, keep the posts where the author continues their own thread.

At least one username, keyword or post link is required.

Speed and cost

Measured: 188 posts from 8 profiles, 5 keyword searches and 3 post links in 27 seconds, every request answered first time. The actor reads Threads directly, without a proxy, which keeps runs fast and very cheap — schedule it as often as you like.

You only pay for posts you get

Profiles that don't exist, private profiles, searches with no results and deleted posts are never written to the dataset, so they are never charged. They are listed with the reason in RUN_SUMMARY.failed.

Limitations

• Logged-out visitors see a limited slice of Threads, and so does this actor: a profile shows its latest few threads (not the full history), a search shows its top ~25 posts, and a post shows its top replies. For a complete history of an account, run the actor on a schedule and keep every run's results.
• View counts are not included — Threads does not show them to visitors.
• Private profiles are not visible without following them.

Legal

Reads only what Threads shows any visitor without logging in. You are responsible for how you use the data.