Bluesky Scraper - Profile, Posts, Followers & Threads

Extract Bluesky profile info, posts, followers, following, and full reply threads by handle. No login, no auth, no proxies needed — reads Bluesky's own public AT Protocol API directly (public.api.bsky.app), the same read-only access Bluesky's own apps use for logged-out browsing.

Quick start

Run with zero configuration and it returns real data (Bluesky's own official account) by default:

python main.py

Or set an input:

{ "handle": "jay.bsky.team", "mode": "profile_and_posts", "limit": 20 }

Modes

mode	returns
profile_and_posts (default)	Profile info + recent posts in one call
profile_only	Just profile info
followers	Follower list
follows	Following list
thread	A post plus its replies (needs postUri)

Why this is safe to run at scale

Every endpoint here (getProfile, getAuthorFeed, getFollowers, getFollows, getPostThread) is a fully public, unauthenticated AT Protocol read — verified live 2026-08-03 against public.api.bsky.app. This is not scraping HTML or defeating any anti-bot measure; it's the exact same data Bluesky's App View serves to anyone, logged out. searchPosts was deliberately left out — it returned a 403 in testing and needs an authenticated session, unlike everything else here.

Verified

test_live.py checks all 5 functions against the real live API before anything gets built on top of them. Also run end-to-end through Apify's local dev runtime (python main.py with zero input) — confirmed a real, useful dataset gets pushed on the very first run with no configuration.

Pricing

Configured for Apify's Pay-Per-Event model (~$0.003-0.005/result) rather than a monthly subscription or free tier — zero friction to try, no payment infra to build, and free listings reportedly signal "hobby project" rather than a maintained tool.

Known competition

There are 10+ existing Bluesky scrapers on Apify already — this is not a zero-competition niche, and shouldn't be sold as one. What's different here: a genuinely zero-config working default, a clean five-endpoint scope (no attempt to also do search, which needs auth), and a legal profile that's about as clean as scraping gets (fully public API, no ToS wall, no anti-bot to route around, Bluesky's own docs actively support this kind of access).