Bluesky Profile Posts Scraper

Collect public Bluesky profile posts and save each publication as a separate dataset row. Bluesky Profile Posts Scraper accepts complete handles, profile links or account identifiers and returns text, dates, authors, visible engagement, images, videos, links and conversation references.

Use a known account as your starting point for content analysis, publication archives and repeated editorial checks. Choose the public profile tab you want and a page budget for each account.

At a glance	Details
📥 Input	Add complete Bluesky handles such as bsky.app or washingtonpost.com, @handles, DIDs, or canonical https://bsky.app/profile/handle URLs.
📤 Output	One unique publication
📄 Pagination	Up to 30 source items per requested page
🔐 Login required	No
⚡ Processing	Up to 5 HTTP requests concurrently; up to 5 attempts for temporary failures
💾 Delivery	One Apify dataset view with individual result rows, flattened object columns and complete nested JSON

What the Bluesky profile posts scraper collects 📝

The Actor reads the public profile feed selected by your mode. Post text, author identity, reply references and media stay together in each record, so you can compare publications without unpacking a page-sized array.

Data can include:

• Post URI, content identifier, canonical URL, text, publication and indexing dates
• Visible like, reply, repost, quote and bookmark counts when exposed
• Author handle, name, biography, avatar, verification and optional profile counts
• Image variants, video playlist and thumbnail links, external link previews and quoted-post context
• Languages, hashtags, mentioned account identifiers, links, reply root and parent references
• Pinned and repost context when present in the returned feed

Mode	Input	Result	Best for
posts	A profile	Posts and the author’s own threads	A standard profile content sample
replies	A profile	Posts and replies, as in the Replies tab	Conversation participation research
media	A profile	The public media tab	Image and media inventories
video	A profile	The public video tab	Video publication research

How to collect profile posts from Bluesky 🚀

1. Enter one or more supported inputs.
2. Choose a mode if available and set the page count for each input.
3. Start the Actor.
4. Open the Posts dataset view and review individual result rows.
5. Export the dataset or retrieve it from your application.

{
  "inputList": [
    "bsky.app"
  ],
  "mode": "posts",
  "pagesPerProfile": 1
}

pagesPerProfile counts feed pages for each account. All four modes request up to 30 source items per page. The replies mode includes ordinary posts as well as replies; it is not a reply-only filter. The profile tab can contain reposts or pinned entries according to Bluesky’s response.

Bluesky profile posts output fields 📦

Field	Type	Meaning
input	string	Original submitted input, retained on success and failure.
status	string	success or failed.
uri	string or null	Stable AT publication URI.
cid	string or null	Content identifier for this version of the post.
post_url	string or null	Canonical URL of the returned publication.
text	string or null	Publication text.
created_at	string or null	Creation or publication time returned by the source.
indexed_at	string or null	Time the source indexed the record.
like_count	integer or null	Visible likes on the returned publication.
repost_count	integer or null	Visible repost count.
reply_count	integer or null	Visible reply count; replies themselves are not expanded.
quote_count	integer or null	Visible quote count.
bookmark_count	integer or null	Bookmark count only when exposed.
author	object or null	Author profile fields.
languages	array or null	Language tags supplied in the post record.
hashtags	array or null	Hashtags extracted from record tags and facets.
mentioned_dids	array or null	Mentioned account identifiers from text facets.
links	array or null	Link targets from text facets.
reply_parent_uri	string or null	Direct parent publication URI when this is a reply.
reply_root_uri	string or null	Root publication URI when this is a reply.
labels	array or null	Source label values.
images	array or null	Ordered image metadata: fullsize and thumbnail URLs, alt text and dimensions.
video	object or null	Video playlist URL, thumbnail, content identifier, alt text and dimensions.
external	object or null	External link preview URL, title, description and thumbnail.
quote	object or null	Embedded quoted-record reference, availability and exposed content.
is_pinned	boolean or null	Whether the feed response marks this entry as pinned.
is_repost	boolean or null	Whether the feed entry is a repost.
reposted_at	string or null	Feed-entry repost time when supplied.
reposted_by	object or null	Reposting profile when supplied by a feed entry.
mode	string or null	Selected collection mode.
source_profile_did	string or null	Identifier of the requested source profile.
source_profile_handle	string or null	Resolved handle of the requested source profile.
source_profile_url	string or null	Canonical URL of the requested source profile.

Nested objects keep the following fields. They appear as flattened columns in the single Store view; arrays stay together in their parent row:

• author: did, handle, display_name, profile_url, description, avatar_url, banner_url, pronouns, created_at, indexed_at, followers_count, following_count, posts_count, verified_status, trusted_verifier_status, labels.
• images[]: fullsize_url, thumbnail_url, alt, width, height.
• video: playlist_url, thumbnail_url, cid, alt, width, height.
• external: url, title, description, thumbnail_url.
• quote: uri, cid, post_url, text, created_at, author, available.
• quote.author: did, handle, display_name, profile_url, description, avatar_url, banner_url, pronouns, created_at, indexed_at, followers_count, following_count, posts_count, verified_status, trusted_verifier_status, labels.
• reposted_by: did, handle, display_name, profile_url, description, avatar_url, banner_url, pronouns, created_at, indexed_at, followers_count, following_count, posts_count, verified_status, trusted_verifier_status, labels.

The successful examples below use normalized public response data. Values are snapshots rather than promises of current content or counts; every top-level output key is included.

Complete representative successful result:

{
  "input": "bsky.app",
  "status": "success",
  "mode": "posts",
  "source_profile_did": "did:plc:z72i7hdynmk6r22z27h6tvur",
  "source_profile_handle": "bsky.app",
  "source_profile_url": "https://bsky.app/profile/did:plc:z72i7hdynmk6r22z27h6tvur",
  "uri": "at://did:plc:z72i7hdynmk6r22z27h6tvur/app.bsky.feed.post/3l6oveex3ii2l",
  "cid": "bafyreicnt42y6vo6pfpvyro234ac4o6ijug6adwwrh7awflgrqlt4zibxq",
  "post_url": "https://bsky.app/profile/did:plc:z72i7hdynmk6r22z27h6tvur/post/3l6oveex3ii2l",
  "text": "👋  Bluesky is an open social network that gives creators independence from platforms, developers the freedom to build, and users a choice in their experience. We're so excited to have you here! \n\nWe share Bluesky updates & news from this account. A quick orientation thread: 🧵✨",
  "created_at": "2024-10-17T07:06:51.491Z",
  "indexed_at": "2024-10-17T07:06:51.491Z",
  "author": {
    "did": "did:plc:z72i7hdynmk6r22z27h6tvur",
    "handle": "bsky.app",
    "profile_url": "https://bsky.app/profile/did:plc:z72i7hdynmk6r22z27h6tvur",
    "labels": [],
    "verified_status": "none",
    "trusted_verifier_status": "valid",
    "display_name": "Bluesky",
    "description": null,
    "avatar_url": "https://cdn.bsky.app/img/avatar/plain/did:plc:z72i7hdynmk6r22z27h6tvur/bafkreihwihm6kpd6zuwhhlro75p5qks5qtrcu55jp3gddbfjsieiv7wuka",
    "banner_url": null,
    "pronouns": null,
    "created_at": "2023-04-12T04:53:57.057Z",
    "indexed_at": null,
    "followers_count": null,
    "following_count": null,
    "posts_count": null
  },
  "languages": [
    "en"
  ],
  "hashtags": [],
  "mentioned_dids": [],
  "links": [],
  "labels": [],
  "reply_parent_uri": null,
  "reply_root_uri": null,
  "is_pinned": true,
  "is_repost": false,
  "reposted_at": null,
  "reposted_by": null,
  "images": [],
  "video": null,
  "external": null,
  "quote": null,
  "like_count": 63684,
  "repost_count": 9529,
  "reply_count": 8579,
  "quote_count": 708,
  "bookmark_count": 250
}

Complete failed dataset item:

{
  "input": "   ",
  "status": "failed",
  "uri": null,
  "cid": null,
  "post_url": null,
  "text": null,
  "created_at": null,
  "indexed_at": null,
  "like_count": null,
  "repost_count": null,
  "reply_count": null,
  "quote_count": null,
  "bookmark_count": null,
  "author": null,
  "languages": null,
  "hashtags": null,
  "mentioned_dids": null,
  "links": null,
  "reply_parent_uri": null,
  "reply_root_uri": null,
  "labels": null,
  "images": null,
  "video": null,
  "external": null,
  "quote": null,
  "is_pinned": null,
  "is_repost": null,
  "reposted_at": null,
  "reposted_by": null,
  "mode": null,
  "source_profile_did": null,
  "source_profile_handle": null,
  "source_profile_url": null
}

Each failed row preserves input, uses status: failed, and sets every other top-level field to null. The reason is written to the run log. Successful rows may contain null optional fields or empty arrays when the source does not supply a value. The dataset is not split into separate tables for media, authors, modes or failures.

Input and pagination settings ⚙️

Parameter	Type	Required	Default	Rules
inputList	array of strings	Yes	None	Add complete Bluesky handles such as bsky.app or washingtonpost.com, @handles, DIDs, or canonical https://bsky.app/profile/handle URLs.
mode	string	No	posts	Choose the source results to collect. Each mode uses the same single dataset view. Allowed: posts, replies, media, video.
pagesPerProfile	integer	No	1	Number of result pages to attempt per input. Each page requests up to 30 items; actual public rows can be fewer. Bootstrap requests do not count as pages. Stops when the cursor ends or repeats. Minimum 1.

Use a complete handle such as bsky.app, @bsky.app, a supported did:plc: or did:web: identifier, or https://bsky.app/profile/bsky.app. A bare local name without a domain is invalid. Record links such as /post/… are not profile inputs.

For supported website URLs, use the exact https://bsky.app host and canonical path without a query string or fragment. URLs from other hosts and unsupported record types are rejected. Handles are normalized; duplicate normalized inputs in the same mode are processed once. Different source inputs retain their own results, while repeated record identifiers within one input’s pagination are removed.

The first result request counts as page 1; resolving a handle or validating source metadata does not consume a result page. Pagination stops at the requested page count, a missing cursor or a repeated cursor. A short, empty or fully filtered page can still continue when it includes a usable next cursor. No exact result total is guaranteed.

Bluesky profile posts use cases 🎯

Editorial and brand monitoring

Collect selected account feeds and compare publication dates, wording, media choices and visible response counts. Schedule later runs when you need fresh snapshots.

Creator content research

Compare posts, media and videos using the same result schema. Author and source-profile fields let you group the output even when a feed entry is a repost.

Content archives and analysis

Export post text, links, identifiers and media metadata to your database. Use the records for your own search, classification, reporting or human review.

Pricing and saved-result behavior 💰

See the Actor’s Pricing tab for the active charging model and current rate. Store settings are separate from this local implementation, so this README does not state an unverified fixed price or runtime.

Under dataset-item pricing:

• Every unique record saved as success is one result for that source input. Nested author, media or profile fields do not become separate rows.
• An invalid or unavailable input, an exhausted temporary failure, or an input with no publicly available results within its page budget can save one failed row.
• Automatic retry attempts do not create extra dataset rows by themselves.
• A normal empty continuation after earlier successes creates no extra result row.
• More requested pages can produce more saved rows. Check the active listing for how saved failed rows are billed; they are not assumed to be free.

The final total depends on source availability, duplicate removal and the chosen page budget. Start with a small run and check its actual usage before selecting a larger budget.

Bluesky profile posts API 🔌

Replace $ACTOR_ID with the identifier from this Actor’s API tab and $APIFY_TOKEN with your Apify token.

curl -X POST "https://api.apify.com/v2/acts/$ACTOR_ID/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"inputList":["bsky.app"],"mode":"posts","pagesPerProfile":1}'

Retrieve the default dataset through the Apify API or download JSON, CSV, Excel, XML and other formats available in the Console. Schedules, completion webhooks and Apify integrations can connect results to Google Sheets, Make, Zapier, n8n, cloud storage or your own backend. These are platform connection options, not integrations preconfigured by this Actor.

Reliability, retries, and public-data limits ⚠️

The Actor uses pure HTTP collection with up to five requests concurrently. It does not launch a browser. Invalid syntax and confirmed missing, removed or inaccessible targets stop without unnecessary retries. Temporary network and proxy failures, timeouts, blocks, malformed responses, throttling and server errors are retried up to five total attempts.

Earlier successful pages remain saved if a later page fails. If no publicly available rows are found before the source ends or the selected page budget is reached, one failed row records that input; this does not mean the underlying account or collection necessarily does not exist. A later empty page after successes is normal exhaustion. A later request that exhausts retries can append a failed row while preserving earlier results.

Bluesky controls public availability and returned fields. Objects restricted from unauthenticated viewing are excluded; a restricted target produces a failed row. Deleted, suspended, unavailable or otherwise restricted records may be missing. Optional counts are not inferred from an incomplete sample, and media links may change or expire.

The Actor collects the selected account tab, not a complete conversation tree. Reply references identify related posts but do not trigger comment extraction. Media links are returned as supplied; the Actor does not download files.

Three consecutive result pages were verified for every paginated mode in HTTP runs of the packaged Actor on September 11, 2026. These checks demonstrate working continuation on the tested sources and do not guarantee future source availability.

One invalid string inside an otherwise valid input list does not stop other inputs. A configuration that fails the input schema logs an input warning and exits without starting requests: for example, a list containing a number instead of a string, an unsupported mode or an invalid page-count type. Infrastructure failures such as startup errors, unavailable dataset storage or an unrecoverable result-save error can still stop the whole run. Result-save failures are not retried as fresh scraping requests.

Frequently asked questions ❓

Does replies mode return only replies?

No. It follows the public Replies tab, which contains posts and replies. Use reply_parent_uri or reply_root_uri to select reply records after export.

Do video rows include downloadable MP4 files?

The Actor returns the video playlist and thumbnail URLs exposed in the response. It does not convert playlists into MP4 files or download media.

Can I choose an exact number of posts?

You choose pages, with up to 30 source items per page. Overlap, unavailable entries and source filtering can reduce the number of saved rows.

Does it require Bluesky login or cookies?

No Bluesky login, password, session cookie or account token is accepted or required. The Actor uses HTTP requests used by the public website and respects restrictions on unauthenticated access.

What happens to invalid or unavailable inputs?

Invalid individual inputs are saved as failed rows without an HTTP request. Confirmed missing, removed or restricted targets also become failed rows without unnecessary retries. Temporary failures are retried up to five total attempts. Other inputs and already saved pages remain available.

Can I export results or automate collection?

Yes. Use the Apify dataset to download JSON, CSV, Excel, XML or other supported formats, or retrieve records through its API. Apify schedules and webhooks can connect repeated runs to your own workflow.

Support, responsible use, and related actors 🛟

For a reproducible problem, open an issue in the Actor’s Issues tab. Include the run ID, approximate time, mode if relevant, page count, safe public input, expected result and actual result. Do not share access tokens, proxy credentials or other secrets.

Use public data responsibly and follow applicable privacy, copyright, contractual and platform requirements before storing, combining or redistributing collected information.