Threads Scraper

Fast HTTP-only scraper for public Meta Threads data. It extracts profiles, user posts, replies, search users, search threads, and engagement metrics without launching a browser.

Supported Actions

Action	Input	Output
getProfiles	usernames	Profile metadata for each user
getUserPosts	usernames, optional sessionId	Posts/threads from each user
getReplies	postUrls	Replies or the main post for each URL
searchUsers	searchQuery	Matching public usernames
searchThreads	searchQuery	Matching public threads found in search HTML

Input

{
    "action": "getUserPosts",
    "usernames": ["zuck", "mosseri"],
    "maxResults": 50,
    "concurrency": 12,
    "proxy": {
        "useApifyProxy": true,
        "apifyProxyGroups": ["RESIDENTIAL"]
    }
}

For deeper user-post pagination, provide an Instagram/Threads sessionid cookie:

{
    "action": "getUserPosts",
    "usernames": ["zuck"],
    "maxResults": 200,
    "sessionId": "YOUR_SESSIONID_COOKIE"
}

Fields

Field	Type	Required	Description
action	enum	Yes	getProfiles, getUserPosts, getReplies, searchUsers, or searchThreads
usernames	string[]	For profiles/posts	Threads usernames, with or without @
postUrls	string[]	For replies	Full Threads post URLs from threads.net or threads.com
searchQuery	string	For search	Search term
maxResults	integer	No	Max results per user/query. Default: 50, max: 1000
concurrency	integer	No	Parallel targets to process. Default: 12, max: 25
sessionId	string	No	Optional sessionid cookie for authenticated mobile API pagination
proxy	object	No	Apify proxy or custom proxy URLs. Residential proxies are recommended

Session ID

searchThreads, searchUsers, and getProfiles are fully public and need no Session ID. You only need one for getReplies and for deeper getUserPosts pagination (beyond ~25 posts).

When you do need it, paste the sessionid cookie value (not the whole Cookie header) into the sessionId input.

How to get your Session ID

1. In a desktop browser, log in to https://www.threads.com (or https://www.instagram.com — they share the cookie).
2. Open DevTools (F12 or right-click then Inspect).
3. Go to the Application tab (Chrome/Edge) or Storage tab (Firefox).
4. In the sidebar, expand Cookies and select the site (e.g. https://www.threads.com).
5. Find the row named sessionid and copy its Value. It looks like 12345678%3AAbCdEf...%3A12%3A....
6. Paste that value into the sessionId field.

Use a throwaway account, not your main one. Meta can flag, rate-limit, or ban accounts used for scraping. Create a dedicated secondary Threads/Instagram account and use its sessionid. Only use a session you are authorized to use. The cookie grants account access, so treat it like a password — the input is stored as a secret.

Pagination

Public, logged-out scraping is limited by Threads. The logged-out search page returns a single page (~10-25 results) with no working public next-page request, and the logged-out GraphQL/profile paths hit a login wall after ~25 posts.

When sessionId is provided, the scraper uses the authenticated mobile API and paginates until maxResults, the API stops returning pages, or rate limiting/session expiry occurs.

Output Examples

Profile

{
    "id": "314216",
    "username": "zuck",
    "fullName": "Mark Zuckerberg",
    "bio": "CEO of Meta",
    "followers": 3200000,
    "following": 523,
    "isVerified": true,
    "profilePicUrl": "https://...",
    "threadCount": 1234,
    "url": "https://www.threads.net/@zuck",
    "scrapedAt": "2026-06-26T10:30:00.000Z"
}

Post

{
    "id": "3456789012345678",
    "shortcode": "CuXyz123",
    "text": "Post content here...",
    "timestamp": "2026-06-26T08:00:00.000Z",
    "likes": 12500,
    "replies": 340,
    "reposts": 890,
    "author": {
        "username": "zuck",
        "fullName": "Mark Zuckerberg",
        "profilePicUrl": "https://...",
        "isVerified": true
    },
    "mediaUrls": ["https://..."],
    "url": "https://www.threads.net/@zuck/post/CuXyz123",
    "scrapedAt": "2026-06-26T10:30:00.000Z"
}

Development

bun install
bun test
bun run typecheck
bun run lint

Notes

• Residential proxies are recommended because Meta frequently blocks datacenter IPs.
• The scraper retries rate limits and transient server errors with exponential backoff.
• Private accounts and login-only content are not available in public mode.
• Threads internal endpoints can change without notice; keep parser tests current when response shapes change.