Threads Search Scraper API: Meta Threads Posts & Replies

Search Threads by keyword and get every public post Threads shows for it: text, author, likes, replies, reposts, quotes, shares, images, videos, links and post date. Add profiles to get followers, bio and links plus their latest threads. Paste post links to get a post with its replies.

No Threads or Instagram account, no cookies of yours, no browser. $2 per 1,000 posts.

What you can use it for

• Brand and competitor monitoring: schedule a run for your brand, product and competitor names and see what people post about them on Threads.
• Trend research: what is being said about a topic right now, with engagement numbers to sort by.
• Influencer research: who posts about your niche, with their follower counts from a profile lookup.
• Social listening for AI agents: give an agent or LLM pipeline fresh Threads posts through the API or the Apify MCP server.

Official Threads API keyword search vs this Actor

Meta's Threads API has a keyword_search endpoint, but searching other people's public posts needs a Meta developer app, a user access token and the threads_keyword_search permission, which Meta grants only after app review. Until it is approved, the endpoint searches only the posts of the account that logged in. Once approved, each user can send up to 2,200 searches per 24 hours (Meta's documentation).

	Threads API keyword_search	This Actor
Setup	Meta app, OAuth login, threads_basic and threads_keyword_search permissions, app review	an Apify token
Whose posts	your own until app review is passed, then public posts	public posts, from the first run
Output	Meta's fields for each post	text, author, likes, replies, reposts, quotes, shares, media and links on every post
Price	free, within Meta's limits	$2 per 1,000 posts

If you already have an approved Meta app, use the official API. If you need public Threads posts by keyword this afternoon, from Python or JavaScript, without app review, use this Actor.

How to use it

1. Put one keyword or phrase per line in Search keywords.
2. Optional: add Profiles (@name or profile URLs). Set Posts per profile to 0 for profile data only.
3. Optional: add Post URLs to get those posts and their replies.
4. Run it and download the results as JSON, CSV or Excel, or read them through the API.

Input example

{
  "searchQueries": ["web scraping", "openai", "nike running"],
  "usernames": ["@mosseri"]
}

Post URLs and replies

{
  "postUrls": ["https://www.threads.com/@zuck/post/DTa3-B1EbTp"],
  "maxRepliesPerPost": 50
}

Each post link gives one row for the post (source: "post") and then its replies (source: "replies"), with the reply's author, text, likes, replyToUsername, replyToPostCode and positionInThread (0 is a direct reply, 1 and up are replies under it). Threads shows a logged-out visitor the first 7 to 11 reply threads of a post, usually 9 or 10 replies; that is what you get, with no login. Each post and each reply is one post result ($0.002).

Date window and author filter

postedAfter and postedBefore keep only keyword and profile posts published inside the window: 2026-10-01, an ISO timestamp, or relative such as 7 days. Add fromUsername (for example meta) to keep one author's posts from a keyword search. Posts outside are skipped and not charged. Inputs written for other Threads scrapers keep their filters here (start_date / end_date, after / before / from).

{
  "searchQueries": ["claude code"],
  "postedAfter": "7 days",
  "maxPostsPerQuery": 50
}

The window filters the posts Threads shows a logged-out visitor (about 55 per keyword across the top, recent and tag tabs; a profile's latest posts), so it narrows that set rather than reaching further back.

Profiles only (Threads profile scraper)

{
  "usernames": ["@zuck", "mosseri", "https://www.threads.com/@natgeo"],
  "maxPostsPerProfile": 0
}

Each username gives one profile row: username, full name, bio, followers, verified badge, bio links and profile picture. With maxPostsPerProfile above 0 you also get that many posts from the profile's latest threads (Threads shows a logged-out visitor about 4). Profile only costs $2 per 1,000 profiles.

Switching from another Threads scraper? Its input runs here as is: searchQuery, keywords, maxPosts, max_posts, maxItemsPerKeyword, sort: "recent", sortByRecent and usernames are all accepted, and so are post and profile link lists in startUrls, post_urls or input, maxReplies, max_replies, includeReplies, include_nested_replies, action with targets, afterDate / beforeDate, maxThreads (posts per profile) and mode: "profile" (profile rows only).

Output

One item per post:

{
  "type": "post",
  "id": "3570811367353526702",
  "code": "DGOEgUHCKmu",
  "url": "https://www.threads.com/@theaiagents/post/DGOEgUHCKmu",
  "text": "Web scraping is dead.\n\nNo more messy code.\n\nNo more Selenium nightmares. ...",
  "publishedAt": "2025-02-18T15:52:34.000Z",
  "username": "theaiagents",
  "fullName": "Zahir Aftab | Ai Agents & Automation",
  "isVerified": true,
  "authorId": "66989702635",
  "likes": 1166,
  "replies": 53,
  "reposts": 131,
  "quotes": 8,
  "shares": 347,
  "mediaType": "video",
  "imageUrls": ["https://scontent-ord5-1.cdninstagram.com/v/t51.71878-15/..."],
  "videoUrls": ["https://scontent-ord5-2.cdninstagram.com/o1/v/t16/....mp4"],
  "linkUrl": null,
  "mentions": [],
  "topic": null,
  "isReply": false,
  "threadId": "3570811367353526702",
  "positionInThread": 0,
  "source": "search",
  "searchQuery": "web scraping",
  "searchTab": "top",
  "position": 1,
  "scrapedAt": "2026-10-03T17:19:20.318Z"
}

And one item per profile:

{
  "type": "profile",
  "username": "mosseri",
  "url": "https://www.threads.com/@mosseri",
  "fullName": "Adam Mosseri",
  "biography": "Father of three boys, married to an amazing woman, head of Instagram and supporter of Threads.",
  "followers": 1046062,
  "isVerified": true,
  "isPrivate": false,
  "bioLinks": ["http://mosseri.net/"]
}

• mediaType is text, image, video or carousel. topic is the Threads topic tag on the post, if any.
• A search result can be a short thread: threadId and positionInThread (0 for the first post) keep them together.
• searchTab says which Threads search view the post came from (top, recent or tags).
• matchedBy and fallbackQuery appear only on posts found by the phrase fallback (see Limits).

The run summary (posts and profiles delivered, keywords without posts, profiles not found, failures) is saved as OUTPUT in the run's key-value store. Results are in the dataset's Posts and Profiles views.

Pricing

Pay per result, no subscription:

Event	Price
Post	$0.002 ($2 per 1,000 posts)
Profile	$0.002
Actor start	$0.00005

A keyword with 50 posts costs $0.10. 100 keywords a day cost about $10. Keywords that find nothing cost nothing. Bronze, Silver and Gold Apify plans get 5%, 10% and 15% off. Proxy and compute are included.

Use it from code

API (cURL)

curl -X POST "https://api.apify.com/v2/acts/sauliusautomatesit~threads-search-api/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchQueries": ["web scraping"]}'

Python

$pip install apify-client

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("sauliusautomatesit/threads-search-api").call(run_input={
    "searchQueries": ["openai"],
    "maxPostsPerQuery": 30,
})
posts = [p for p in client.dataset(run.default_dataset_id).iterate_items() if p.get("type") == "post"]
for p in sorted(posts, key=lambda p: p["likes"] or 0, reverse=True)[:10]:
    print(p["likes"], p["username"], (p["text"] or "")[:80])

run.default_dataset_id is for apify-client 3.x (the current release); on 2.x write run["defaultDatasetId"]. Pass max_total_charge_usd=1 to .call() to cap a run at $1.

JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_TOKEN' });
const run = await client.actor('sauliusautomatesit/threads-search-api').call({ searchQueries: ['web scraping'] });
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.map((p) => `${p.likes} ${p.url}`));

MCP (Claude, Cursor, any MCP client)

Add https://mcp.apify.com/?tools=sauliusautomatesit/threads-search-api as a remote MCP server and your agent can search Threads as a tool.

Limits, honestly

• Threads shows visitors who are not logged in about 15 to 20 posts per search view, with no next page. The Actor asks three views (top, recent, tags) and merges them, so a keyword usually returns 30 to 60 posts. For more, use several related keywords. Scheduled runs pick up new posts as Threads surfaces them.
• Results are what Threads search returns, and Threads matches loosely: a short or unusual keyword can bring posts that only resemble it.
• Threads answers a logged-out visitor only for phrases it already knows. coffee shop london and кофейня Алматы return posts, coffee sale, Алматы кофейня and продам кофейню return none. When a phrase of 2 or more words returns nothing, the Actor tries its other word orders, then searches each word alone and keeps only posts that contain every word (Russian and English word endings allowed). A word Threads has nothing for as typed is also searched in its base form (стоматолога finds nothing, стоматолог does; dentists becomes dentist), for single word keywords too. Request words such as ищу, посоветуйте, looking for and recommend are not required in the post, so ищу стоматолога returns posts about dentists. Those rows carry matchedBy (wordOrder, allWords or baseForm) and fallbackQuery. This finds some posts for most such phrases, not all; short common words in their dictionary form work best. Set exactPhraseOnly to true to turn it off.
• Profiles return the profile and the 4 latest threads Threads shows on the profile page. Private profiles show the profile only.
• Post URLs return the post and the first replies Threads shows without a login (usually 9 or 10). Posts with hundreds of replies still give only that first page.
• Public posts only. No login, so no private accounts, likes lists or follower lists.

Related Actors

• Telegram Channel Scraper: messages, views and reactions from public Telegram channels.
• Bing Search API: Bing web results for any list of queries.
• Google Trends API: search interest over time, by region, related and rising queries.

FAQ

Do I need a Threads or Instagram account? No. The Actor reads what Threads shows any visitor.

Do I need Meta app review or the threads_keyword_search permission? No. Those are for Meta's own Threads API. This Actor needs only an Apify token.

Is there a free Threads API for searching posts? Meta's API is free once your app passes review. Here, Apify's free plan's $5 of monthly credit covers about 2,500 posts.

Can I monitor a brand every day? Yes. Save the input as a task and schedule it; each run returns the posts Threads shows at that moment, with fresh engagement numbers.

How fast is it? A few seconds per keyword, 5 keywords in parallel by default (up to 20).

Is it legal? It collects publicly visible posts. You are responsible for how you use the data and for complying with the laws that apply to you, including data protection laws for personal data.