Bluesky Scraper: posts, search, profiles, followers and threads

Collect public data from Bluesky through the official AT Protocol API. Give it a list of handles, search terms or post URLs and it returns clean JSON rows: posts with full engagement counts, profiles with follower and post totals, follower and following lists, user search results, or a complete reply thread.

Because it reads Bluesky's official API rather than scraping HTML, results are fast and stable. No login or cookies are needed for anything except keyword post search, which Bluesky itself gates behind login (a revocable app password of your own).

What it does

Operation	You provide	You get
Posts	handles or profile URLs	Each user's posts with likes, reposts, replies, quotes, images, link cards
Profiles	handles or profile URLs	Bio, follower count, following count, post count per user
Followers	handles	The accounts following each user
Following	handles	The accounts each user follows
Search users	keywords	Matching accounts with profile data
Search posts	keywords + your app password	Latest posts matching each keyword, with engagement counts
Thread	post URLs	The post and every reply beneath it, flattened

Typical uses: social listening and brand monitoring, influencer research, audience analysis, tracking what a set of accounts posts over time (run it on a schedule and diff by post URI), and feeding Bluesky data to AI agents. The Actor works from Apify's MCP server, so agents in Claude, Cursor and similar tools can call it directly.

Input

{
    "operation": "posts",
    "targets": ["bsky.app", "https://bsky.app/profile/atproto.com"],
    "maxItemsPerTarget": 100,
    "includeReplies": false,
    "includeReposts": false
}

Targets are flexible: handles (name.bsky.social), custom-domain handles (bsky.app), DIDs, or profile URLs all work. For the thread operation, paste a post URL straight from your browser.

Output

One row per item. Post rows:

{
    "type": "post",
    "uri": "at://did:plc:z72i7hdynmk6r22z27h6tvur/app.bsky.feed.post/3l6oveex3ii2l",
    "url": "https://bsky.app/profile/bsky.app/post/3l6oveex3ii2l",
    "handle": "bsky.app",
    "displayName": "Bluesky",
    "text": "…",
    "createdAt": "2026-07-01T18:04:12.000Z",
    "likeCount": 4120,
    "repostCount": 388,
    "replyCount": 214,
    "quoteCount": 77,
    "isReply": false,
    "images": ["https://cdn.bsky.app/img/feed_fullsize/..."],
    "externalLink": null,
    "source": "bsky.app"
}

User rows carry handle, displayName, description, followersCount, followsCount, postsCount and avatar.

Export as JSON, CSV, Excel or RSS, or read the dataset through the Apify API from your own code.

Pricing

Pay per event: a small fee when a run starts, then a fee per row in your results. Duplicate items are filtered out before charging.

Use from AI agents

This Actor works out of the box with AI agents through Apify's MCP server. Connect an agent (Claude, Cursor, or anything MCP-compatible) to Apify and it can call apeye/bluesky-scraper directly: ask "what has this account posted about pricing this month, and who are its most-followed followers?" and the agent picks the operations, runs them and reads the results. Row types and field names are kept flat and predictable so models can consume them without custom parsing.

FAQ

Is this legal? Yes. The Actor reads the official, publicly documented AT Protocol API that serves Bluesky's own app. Only public data is collected, the same data anyone sees without logging in.

How does post search work? Bluesky requires login for post search (their rule, not ours), so the searchPosts operation takes your handle and an app password. App passwords are Bluesky's official mechanism for third-party tools: create one at Settings > App Passwords, revoke it any time, and it never grants access to your main password or account settings. The credential field is stored encrypted and used only to log in. Every other operation needs no credentials at all.

How fresh is the data? It's read live from the API at run time. Engagement counts are the values at the moment of the run.

What about rate limits? The Actor paces itself within Bluesky's public API limits and retries transient errors automatically.

Support

If something breaks or you need another operation (lists, likes of a post, feeds), open an issue on the Issues tab. I check most days and usually reply within a day or two.