Bluesky Scraper — Posts, Profiles, Followers, Threads & Search for AI Agents & Data Pipelines

The all-in-one Bluesky scraper for AI agents and data pipelines: search posts, fetch profiles, export author feeds, followers/following graphs, full conversation threads, and user search — and get back typed JSON rows validated against a published output schema (documented below). Built on the official public AT Protocol AppView API: no login, no cookies, no proxies, data is near-realtime, and every run's output is exportable as JSON, CSV, or Excel.

Why AI agents use this Bluesky scraper

• Deterministic typed output — every dataset row validates against the actor's dataset schema; field names and types never drift between runs or modes.
• Cost-predictable per-event pricing — a fixed price per result plus a start fee ($1 per 1,000 results, cheapest in the Apify Store). maxItems hard-caps spend, so autonomous budgets are safe.
• Zero auth, zero proxies — reads the official public AT Protocol AppView anonymously. No credentials to manage, no account risk, and no hidden proxy costs ever (many scrapers bill proxy traffic on top of results — this one cannot, by construction).
• Clear error semantics — invalid input fails fast with exit code 1 and a human-readable status message; nothing fails silently (details in Error semantics below).
• Rate-limit handling built in — 429/5xx responses retry automatically with exponential backoff; WAF 403s fall back to an alternate official host.

Modes

Mode	What it does	Row shape	Needs
search-posts (default)	Full-text post search with sort, since, until, author, mentions, lang filters	post	queries
profiles	Full profile details (bio, banner, follower/post counts)	profile	targets
author-feed	All posts by given accounts, filterable (no replies / with media / …)	post	targets
followers	Who follows the given accounts	profile	targets
following	Who the given accounts follow	profile	targets
thread	Full conversation thread for given post URLs, flattened to rows	post	postUrls
search-users	Search user accounts by keyword	profile	queries

Input schema

Field	Type	Required	Default	Description
mode	string (enum)	yes	search-posts	One of search-posts, profiles, author-feed, followers, following, thread, search-users.
queries	array of strings	for search-posts, search-users	—	Keywords or phrases to search for. Supports Bluesky search syntax ("exact phrase", from:handle).
targets	array of strings	for profiles, author-feed, followers, following	—	Accounts — handles (bsky.app or @bsky.app), DIDs (did:plc:...), or profile URLs (https://bsky.app/profile/bsky.app).
postUrls	array of strings	for thread	—	Post URLs (https://bsky.app/profile/<handle>/post/<id>) or AT-URIs (at://<did>/app.bsky.feed.post/<id>); returns root + all replies up to 10 levels deep.
maxItems	integer	no	100	Maximum result rows in total across all queries/targets. Controls cost: you are charged per result.
sort	string (enum)	no	latest	search-posts only. latest (newest first) or top (most relevant).
since	string	no	—	search-posts only. Only posts on/after this date, ISO format (2024-11-01 or 2024-11-01T00:00:00Z).
until	string	no	—	search-posts only. Only posts before this date, ISO format.
author	string	no	—	search-posts only. Only posts by this account (handle, DID, or profile URL).
mentions	string	no	—	search-posts only. Only posts mentioning this account (handle, DID, or profile URL).
lang	string	no	—	search-posts only. 2-letter language code (en, ja, pt).
postFilter	string (enum)	no	posts_no_replies	author-feed only. posts_no_replies, posts_with_replies, posts_with_media, or posts_and_author_threads.

Output schema

Rows come in two shapes — post rows (search-posts, author-feed, thread) and profile rows (profiles, followers, following, search-users) — plus a few mode-specific fields. Fields absent from a row's shape are simply omitted.

Field	Type	Nullable	Modes	Description
query	string	no	search-posts, search-users	The search query that produced this row.
target	string	no	author-feed, followers, following	Normalized target account (handle or DID) this row was scraped for.
threadOf	string	no	thread	The input post URL whose thread this row belongs to.
parentUri	string	yes	thread	AT-URI of the parent post; null for the thread root when it has no parent.
depth	integer	no	thread	0 = the requested post, 1 = direct reply, etc.
isRepost	boolean	no	author-feed	Feed entry is a repost by the target account.
isPin	boolean	no	author-feed	Feed entry is the account's pinned post.
uri	string	no	post rows	Canonical AT-URI (at://did/app.bsky.feed.post/rkey).
cid	string	no	post rows	Content identifier of the post record.
url	string	no	all	bsky.app post URL (post rows) or profile URL (profile rows).
author	object	no	post rows	{ handle, displayName, did, avatar } — each string or null.
text	string	no	post rows	Full post text (empty string for text-less posts).
createdAt	string	yes	all	ISO 8601 creation time of the post (post rows) or account (profile rows).
likeCount	integer	no	post rows	Like count (0 when absent).
repostCount	integer	no	post rows	Repost count.
replyCount	integer	no	post rows	Reply count.
quoteCount	integer	no	post rows	Quote-post count.
langs	array of strings	no	post rows	Language codes declared on the post record.
hashtags	array of strings	no	post rows	Hashtags parsed from rich-text facets (no # prefix).
links	array of strings	no	post rows	External link URLs parsed from rich-text facets.
embedType	string	yes	post rows	AT Protocol embed view type (e.g. app.bsky.embed.images#view); null when no embed.
imageUrls	array of strings	no	post rows	Full-size image URLs from image embeds (empty when none).
videoPlaylistUrl	string	yes	post rows	HLS playlist URL for video embeds; null otherwise.
externalUrl	string	yes	post rows	URL of an external-link card embed; only present when the post has one.
did	string	no	profile rows	Decentralized identifier of the account.
handle	string	no	profile rows	Bluesky handle.
displayName	string	yes	profile rows	Display name.
description	string	yes	profile rows	Profile bio.
avatar	string	yes	profile rows	Avatar image URL.
banner	string	yes	profile rows	Banner image URL (full profile views; null in list views).
followersCount	integer	yes	profile rows	Follower count (null in list views that omit counts).
followsCount	integer	yes	profile rows	Following count (null in list views that omit counts).
postsCount	integer	yes	profile rows	Post count (null in list views that omit counts).
indexedAt	string	yes	profile rows	When the AppView last indexed the account.
labels	array of strings	no	profile rows	Moderation label values (e.g. !no-unauthenticated).

Example item — search-posts

{
    "query": "apify",
    "uri": "at://did:plc:3qx3wgjvwbhbr2uvbv5qnot4/app.bsky.feed.post/3mt2dck2a3u2p",
    "cid": "bafyreif5cxxqatw3om7sfto6xhtoovzfq6ewpw6dpcruhavjwv3cvsdgcq",
    "url": "https://bsky.app/profile/pingoru.bsky.social/post/3mt2dck2a3u2p",
    "author": {
        "handle": "pingoru.bsky.social",
        "displayName": "Pingoru.io",
        "did": "did:plc:3qx3wgjvwbhbr2uvbv5qnot4",
        "avatar": "https://cdn.bsky.app/img/avatar/plain/did:plc:.../bafkrei..."
    },
    "text": "Update: Apify marked this resolved at 14:05 UTC.",
    "createdAt": "2026-08-14T14:05:43Z",
    "likeCount": 12,
    "repostCount": 3,
    "replyCount": 1,
    "quoteCount": 0,
    "langs": ["en"],
    "hashtags": ["apify"],
    "links": ["https://status.apify.com"],
    "embedType": "app.bsky.embed.images#view",
    "imageUrls": ["https://cdn.bsky.app/img/feed_fullsize/plain/did:plc:.../bafkrei..."],
    "videoPlaylistUrl": null
}

Example item — profiles

{
    "did": "did:plc:z72i7hdynmk6r22z27h6tvur",
    "handle": "bsky.app",
    "url": "https://bsky.app/profile/bsky.app",
    "displayName": "Bluesky",
    "description": "The official Bluesky account.",
    "avatar": "https://cdn.bsky.app/img/avatar/plain/did:plc:.../bafkrei...",
    "banner": "https://cdn.bsky.app/img/banner/plain/did:plc:.../bafkrei...",
    "followersCount": 3452871,
    "followsCount": 8,
    "postsCount": 312,
    "createdAt": "2023-04-12T04:53:57Z",
    "indexedAt": "2026-08-10T02:11:08Z",
    "labels": []
}

Example item — author-feed

Same shape as search-posts but with target, isRepost, isPin instead of query:

{
    "target": "bsky.app",
    "isRepost": false,
    "isPin": true,
    "uri": "at://did:plc:z72i7hdynmk6r22z27h6tvur/app.bsky.feed.post/3l6oveex3ii2l",
    "cid": "bafyreihx...",
    "url": "https://bsky.app/profile/bsky.app/post/3l6oveex3ii2l",
    "author": { "handle": "bsky.app", "displayName": "Bluesky", "did": "did:plc:z72i7hdynmk6r22z27h6tvur", "avatar": "https://cdn.bsky.app/img/avatar/plain/..." },
    "text": "Welcome to Bluesky!",
    "createdAt": "2024-10-24T17:02:23Z",
    "likeCount": 88231, "repostCount": 12045, "replyCount": 4102, "quoteCount": 903,
    "langs": ["en"], "hashtags": [], "links": [],
    "embedType": null, "imageUrls": [], "videoPlaylistUrl": null
}

Example item — followers / following

Profile shape plus target:

{
    "target": "bsky.app",
    "did": "did:plc:44ybard66vv44zksje25o7dz",
    "handle": "jay.bsky.team",
    "url": "https://bsky.app/profile/jay.bsky.team",
    "displayName": "Jay",
    "description": "CEO of Bluesky.",
    "avatar": "https://cdn.bsky.app/img/avatar/plain/...",
    "banner": null,
    "followersCount": null,
    "followsCount": null,
    "postsCount": null,
    "createdAt": "2023-04-12T04:53:57Z",
    "indexedAt": "2026-08-01T00:00:00Z",
    "labels": []
}

(List views omit counts — expect null for followersCount/followsCount/postsCount/banner here; run profiles mode on the handles if you need counts.)

Example item — thread

Post shape plus threadOf, parentUri, depth:

{
    "threadOf": "https://bsky.app/profile/bsky.app/post/3l6oveex3ii2l",
    "parentUri": "at://did:plc:z72i7hdynmk6r22z27h6tvur/app.bsky.feed.post/3l6oveex3ii2l",
    "depth": 1,
    "uri": "at://did:plc:abc.../app.bsky.feed.post/3l6ovxyz...",
    "cid": "bafyreig...",
    "url": "https://bsky.app/profile/someone.bsky.social/post/3l6ovxyz...",
    "author": { "handle": "someone.bsky.social", "displayName": "Someone", "did": "did:plc:abc...", "avatar": null },
    "text": "Congrats on the launch!",
    "createdAt": "2024-10-24T17:10:00Z",
    "likeCount": 5, "repostCount": 0, "replyCount": 0, "quoteCount": 0,
    "langs": ["en"], "hashtags": [], "links": [],
    "embedType": null, "imageUrls": [], "videoPlaylistUrl": null
}

Example item — search-users

Profile shape plus query:

{
    "query": "apify",
    "did": "did:plc:xyz...",
    "handle": "apify.com",
    "url": "https://bsky.app/profile/apify.com",
    "displayName": "Apify",
    "description": "Full-stack web scraping platform.",
    "avatar": "https://cdn.bsky.app/img/avatar/plain/...",
    "banner": null,
    "followersCount": null,
    "followsCount": null,
    "postsCount": null,
    "createdAt": "2023-07-01T09:00:00Z",
    "indexedAt": "2026-08-12T10:00:00Z",
    "labels": []
}

Error semantics

• Bad input fails fast. A mode missing its required field (queries, targets, or postUrls), or a malformed post URL in thread mode, aborts the run immediately with exit code 1 and an explanatory status message (e.g. Mode "profiles" requires at least one entry in "targets"). No partial charge beyond the start event.
• Invalid targets fail loudly. A nonexistent handle or invalid parameter returns a permanent 4xx from the API; the run fails with the API's error detail in the message. Rows scraped before the failure are already in the dataset — nothing is silently dropped.
• Transient errors are retried. 429 and 5xx responses get 3 attempts with exponential backoff (1s, 2s); 403 WAF blocks automatically fall back to the alternate official host. Only after all retries fail does the run fail.
• Zero results is a success. A query matching nothing completes normally with an empty dataset (exit 0) — treat "run succeeded, 0 items" as a valid empty answer, not an error to retry.

Use from AI agents (MCP)

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com/?tools=apricot_blackberry/bluesky-all-in-one",
      "headers": { "Authorization": "Bearer <YOUR_APIFY_TOKEN>" }
    }
  }
}

Works in Claude, Cursor, ChatGPT deep research connectors, and any MCP client; the input schema above is the tool's parameter schema.

Use from code

curl (synchronous run, returns dataset items directly):

curl -X POST "https://api.apify.com/v2/acts/apricot_blackberry~bluesky-all-in-one/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"mode":"search-posts","queries":["apify"],"maxItems":100}'

JavaScript (apify-client):

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('apricot_blackberry/bluesky-all-in-one').call({
    mode: 'search-posts',
    queries: ['apify'],
    maxItems: 100,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Python (apify-client):

from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("apricot_blackberry/bluesky-all-in-one").call(run_input={
    "mode": "search-posts",
    "queries": ["apify"],
    "maxItems": 100,
})
items = client.dataset(run["defaultDatasetId"]).list_items().items

Use from automation platforms

• n8n / Make / Zapier — use the native Apify integration and pick this actor by name (bluesky-all-in-one); the run's dataset feeds directly into the next step.
• LangChain / LlamaIndex — call it through the Apify tool wrappers (ApifyActorsTool / ApifyActor loader) with the input JSON shown above.
• Webhooks — Apify webhooks can fire on run completion (e.g. ACTOR.RUN.SUCCEEDED) to hand results to your own pipeline automatically, no polling required.

Pricing

Pay per event — you only pay for what you scrape:

Event	Price	When
Actor start	per run	Once per run
Result	per row	Every dataset row, all modes

That works out to $1 per 1,000 results — no subscription, no proxy fees, no platform-usage surprises, and maxItems caps your spend on every run.

FAQ

Do I need a Bluesky account or app password? No. The actor only reads public data through Bluesky's official public AppView API. Nothing to configure, no account at risk.

Can I export Bluesky followers to CSV? Yes — run the followers mode with your target handle, then download the dataset as CSV from the run's Storage tab (or via the dataset API).

How fresh is the data? Near-realtime. The AppView indexes the AT Protocol firehose continuously; a search-posts run with sort: latest returns posts from seconds ago.

Does it scrape private or deleted content? No. Only publicly visible content served by the official AppView. Blocked, deleted, or private data is not accessible — this keeps the actor compliant and stable.

What happens if I hit a rate limit? The actor retries automatically with exponential backoff. For very large jobs, split targets across multiple runs.

Can I search posts by a specific user within a date range? Yes — use search-posts with author, since, and until set. Or use author-feed to get a user's complete timeline.

Changelog

• 1.1 — Published dataset output schema (typed, validated rows) + agent-first docs: MCP config, curl/JS/Python snippets, error semantics.
• 1.0 — Initial release: 7 modes (search-posts, profiles, author-feed, followers, following, thread, search-users), facet parsing (hashtags/links), media extraction, pay-per-result pricing.