Bluesky Post & Profile Scraper

Scrape Bluesky without an account. Give it handles, post links, custom feeds or a people search, and get back flat rows of posts and profiles — text, engagement counts, media links, timestamps, followers.

It talks to Bluesky's own public API (the AT Protocol AppView), the same service the website uses. No login, no cookies, no browser, no proxy needed.

You are charged per row delivered. Filtered-out rows and failed inputs cost nothing.

What you can scrape

Input field	What you get
Accounts — scrape their posts	The account's timeline: posts, replies, reposts, media posts
Posts	One row per post, or the entire reply thread under it
Accounts — profile details only	Bio, follower/following/post counts, account creation date
People search	Accounts matching a name, handle or bio text
Custom feeds and lists	Posts in any feed generator or moderation/curation list
Followers of / Following of	The follower or following list as profile rows
Engagement on posts	Who liked, reposted or quoted a given post

Mix as many as you like in one run. Accepts handles (jay.bsky.team), profile URLs, post URLs, feed and list URLs, and raw at:// URIs.

Filters

keywords, excludeKeywords, postedWithinDays, minLikes, languages, mediaOnly, excludeReplies, and an author filter for posts-only / media-only timelines. Filters run before billing.

Output

One flat row per post or profile — ready for CSV, Google Sheets or a database.

{
  "type": "post",
  "uri": "at://did:plc:z72i7hdynmk6r22z27h6tvur/app.bsky.feed.post/3l6oveex3ii2l",
  "url": "https://bsky.app/profile/bsky.app/post/3l6oveex3ii2l",
  "text": "…",
  "createdAt": "2026-09-12T14:03:11.000Z",
  "authorHandle": "bsky.app",
  "authorDisplayName": "Bluesky",
  "authorDid": "did:plc:z72i7hdynmk6r22z27h6tvur",
  "likeCount": 1820, "repostCount": 244, "replyCount": 96, "quoteCount": 31,
  "isReply": false, "replyToUri": null, "threadRootUri": null,
  "embedType": "images", "images": ["https://cdn.bsky.app/…"], "videoUrl": null,
  "externalUrl": null, "links": [], "mentions": [], "hashtags": [], "langs": ["en"]
}

Profile rows carry handle, displayName, description, followersCount, followsCount, postsCount, createdAt, avatar, plus the reason they are there (engagement, relation, matchedQuery).

What it cannot do

• No keyword search over posts. Bluesky requires a logged-in account for app.bsky.feed.searchPosts, and this actor does not log in or work around that. Use the people search field to find accounts by keyword, then scrape their posts. To follow a topic, add a custom feed for it.
• No likes of an account. getActorLikes also requires authentication.
• Public data only. No DMs, no follower lists of accounts that hide them, no deleted or blocked-from-you content.
• No posting, liking or following. Read-only.
• Reply threads go 10 levels deep at most, and very large threads are capped by the API itself.
• Engagement lists (likes, reposts) are returned newest-first and are paginated; on a viral post, pulling all of them takes many requests — use maxItemsPerSource to bound it.
• Counts are what Bluesky reports at fetch time; they are not recomputed.

Tips

• maxItems caps the entire run — the safest way to cap what you spend.
• maxConcurrency above 4 can trigger throttling on the public API. The actor retries automatically, but the run will not be faster.
• Rows that failed carry success: false and an error message; the run still finishes with everything else. Those rows are free.

Legal

This actor reads data that Bluesky publishes openly. You are responsible for how you use it, including GDPR/CCPA obligations when the data concerns people in those jurisdictions.