Bluesky Scraper — AT Protocol API, No Login Required

Production-grade Bluesky scraper using the official AT Protocol (public.api.bsky.app). Profile + author-feed scraping needs zero authentication; keyword search uses Bluesky App Password.

Why Bluesky data?

Bluesky has grown from ~10M to ~32M users since the open-public launch. Built on the open AT Protocol, it's one of the few major social networks where public posts and profile data are exposed via a free, official API — no anti-bot, no rate limit on public endpoints for moderate workloads, no HTML drift.

Why this scraper

• ✅ Official AT Protocol — com.atproto.identity.resolveHandle → app.bsky.actor.getProfile → app.bsky.feed.getAuthorFeed. No HTML parsing.
• ✅ No login for profile + author-feed scraping — public endpoints work without credentials.
• ✅ Search support with Bluesky App Password (App Passwords are revocable, not your main password).
• ✅ Cursor-based pagination — automatic, walks the entire timeline up to maxPostsPerSource.
• ✅ Tested in production — 23+ runs on Apify Cloud, no anti-bot incidents.

Output Data — Profile (12 fields)

{
  "_type": "PROFILE",
  "did": "did:plc:z72i7hdynmk6r22z27h6tvur",
  "handle": "bsky.app",
  "displayName": "Bluesky",
  "description": "see what's next",
  "avatar": "https://cdn.bsky.app/img/avatar/...",
  "banner": "https://cdn.bsky.app/img/banner/...",
  "followersCount": 3242017,
  "followsCount": 4,
  "postsCount": 725,
  "createdAt": "2023-04-12T04:53:57.057Z",
  "scrapedAt": "2026-04-29T12:30:00.000Z"
}

Output Data — Post (24 fields)

{
  "_type": "POST",
  "source": "profile:bsky.app",
  "scrapedAt": "2026-04-29T12:30:00.000Z",
  "uri": "at://did:plc:.../app.bsky.feed.post/3kqxxxxxxx",
  "cid": "bafyrei...",
  "author": {
    "did": "did:plc:z72i7hdynmk6r22z27h6tvur",
    "handle": "bsky.app",
    "displayName": "Bluesky"
  },
  "text": "Welcome to Bluesky!",
  "createdAt": "2026-03-14T01:56:08.229Z",
  "indexedAt": "2026-03-14T01:56:09.123Z",
  "likeCount": 843,
  "repostCount": 77,
  "replyCount": 77,
  "quoteCount": 59,
  "hasImages": false,
  "hasVideo": false,
  "hasLink": true,
  "externalLink": "https://bsky.social",
  "externalTitle": "Bluesky Social",
  "labels": [],
  "languages": ["en"],
  "hashtags": ["#bluesky"],
  "mentions": ["did:plc:..."],
  "isReply": false,
  "parentUri": null
}

Field reference (full): _type, source, scrapedAt, uri, cid, author{did, handle, displayName}, text, createdAt, indexedAt, likeCount, repostCount, replyCount, quoteCount, hasImages, hasVideo, hasLink, externalLink, externalTitle, labels, languages, hashtags, mentions, isReply, parentUri.

Use Cases

• Brand monitoring — track mentions of your brand on Bluesky
• Sentiment analysis — analyze public opinion on trending topics
• Influencer research — find Bluesky accounts by follower count + engagement
• AI training data — build NLP datasets from public Bluesky conversations
• Competitive intelligence — monitor competitor handles + their posting cadence
• Migration analysis — track Twitter/X-to-Bluesky cross-posting patterns
• Academic research — study decentralized social networks

Input Parameters

Parameter	Type	Default	Description
handles	Array	[]	Bluesky handles to scrape (e.g., bsky.app, jay.bsky.team)
searchQueries	Array	[]	Keyword search across all of Bluesky (requires credentials)
maxPostsPerSource	Number	100	Max posts collected per handle or per query
includeLikes	Boolean	false	(reserved) — placeholder for future like-feed support
includeReposts	Boolean	false	(reserved) — placeholder for future repost-feed support
blueskyHandle	String	—	Your Bluesky handle (required only for search)
blueskyPassword	String	—	App Password from bsky.app/settings/app-passwords (required only for search)

How It Works

1. Profile + author-feed (no auth): resolve handle → DID → fetch profile → walk author feed via cursor pagination.
2. Keyword search (auth required): com.atproto.server.createSession → bearer token → app.bsky.feed.searchPosts with cursor pagination.

Step-by-Step

1. Open the actor

apify.com/knotless_cadence/bluesky-scraper → "Try for free."

2. Provide handles (no auth)

{
  "handles": ["bsky.app", "jay.bsky.team"],
  "maxPostsPerSource": 50
}

3. Or run a search (App Password)

{
  "searchQueries": ["llm", "rag"],
  "maxPostsPerSource": 100,
  "blueskyHandle": "you.bsky.social",
  "blueskyPassword": "abcd-1234-..."
}

4. Results

Profile JSON (1 record per handle, _type: PROFILE) plus post JSON (_type: POST, full 24-field schema above) emitted to the Apify dataset.

Pricing

• Standard Apify per-run compute pricing — no per-post or per-profile fee.
• No proxy required — public AT Protocol endpoints are not currently anti-bot gated for moderate volume. AT Protocol response sizes are small (~1-3 KB per post), so compute consumption is low.
• Heavier search jobs (auth-required) trade compute for credential rate-limits — see Bluesky's App Password rate guidance if running >1000 search posts per day.

Honest Limitations

• hashtags extraction is ASCII-word regex (/#\w+/g) — \w only matches [A-Za-z0-9_], so non-ASCII hashtags (#искусственныйинтеллект, #人工知能, #café) are NOT captured. If Unicode hashtags matter for your dataset, post-process the text field with a Unicode-aware tokenizer.
• hasImages / hasVideo / hasLink are exclusive (=== on record.embed.$type). A post using app.bsky.embed.recordWithMedia (quote-post with attached media) will have all three flags FALSE — the embed type is recordWithMedia, not images/video/external. Same for app.bsky.embed.record (pure quote posts).
• mentions extracts only f.features[0]?.did — the FIRST feature in each facet, not all features. Mentions inside a single facet beyond the first are silently lost. In practice, Bluesky places one mention per facet, so impact is minimal — but it is not a complete extraction.
• Cursor pagination is best-effort. If feed.cursor is missing or null, iteration stops — even if collected < maxPostsPerSource. The actor returns whatever it managed to collect; it does not retry the cursor request.
• Author-feed endpoint (app.bsky.feed.getAuthorFeed) returns posts AND reposts in chronological order. The actor does NOT filter reposts out by default — isReply is set, but there is no isRepost field in the schema. If you need only original posts, filter on record.text !== '' and absence of repost markers downstream.
• Search requires App Password. The bsky.social auth endpoint is rate-limited per account. Heavy search workloads from a single App Password may receive HTTP 429.
• No retry logic. A transient 5xx during cursor pagination ends the loop early for that handle/query.

Proof of delivery: This Bluesky scraper has 25 lifetime production runs as of May 2026. Author maintains 31 published actors (78 total) and shipped a paid 3-article series in March 2026 ($150, proxy industry). Pilot pricing locked through May 2026.

Sample request? Reply sample to spinov001@gmail.com and we'll send 2 published case-study articles within 24 hours.

Custom scraping — pricing

Need data from a different network or in a custom schema? One-shot pilot tiers:

• Pilot — $97: 1 actor, basic config, 7-day support. Perfect for proof-of-concept.
• Standard — $297: custom actor + Slack/email alerts on results, 30-day support. Most clients.
• Premium — $797: custom actor + dashboard + 90-day support + 1 modification round.

Email spinov001@gmail.com with the source URL + the fields you need. Typical turnaround: 48 hours.

Related actors (verified live)

• Walmart Reviews Scraper — Product reviews to CSV/JSON/Excel, 17 fields per review, bypasses Walmart's 100-review UI cap
• Reddit Discussion Scraper — 82 runs, 20-field post schema
• YouTube Comments Scraper — comment threads + replies
• Trustpilot Review Scraper — 951 production runs
• Google News Scraper — RSS + HTML fallback

---

Proof of work: 31 public actors on Apify Store (78 total in portfolio). Production-tested: Trustpilot 951 / Reddit 82 / Google News 45 / Glassdoor 39 / Email Extractor 107 / Hacker News 27 / Bluesky 25.

More tips: t.me/scraping_ai

---

Honest disclosure

• This actor uses the public Bluesky AT Protocol API — not HTML scraping. Profile + author-feed endpoints are documented as public; search requires user authentication.
• Output fields reflect what extractPostData() and getProfile() actually push. No fabricated fields.
• Bluesky-related actors mentioned in earlier README versions (bluesky-feed-monitor, bluesky-hashtag-tracker, bluesky-profile-batch) are not currently public — only this scraper is live in the Apify Store. Removed dead links.
• Provided App Passwords are sent only to Bluesky's own auth endpoint (bsky.social); the actor does not log or persist them.
• Not affiliated with Bluesky Social PBC or the AT Protocol team.