# Scrapier — Bluesky API > Scrapier endpoints for Bluesky: Bluesky Profile Info, Bluesky Single Posts, Bluesky User Posts. Base URL: https://api.scrapier.io/api/v2 · Auth: header `X-API-Key: ` · Synchronous: HTTP 200 returns {"success": true, "data": {...}} · Paginate by sending data.next_cursor back as `cursor` until it is null · Paths keep the trailing slash. Live parameter source (no auth): https://api.scrapier.io/api/v2/endpoints · Docs: https://docs.scrapier.io/api-reference/bluesky.md ## GET /bluesky-profile-info/ Bluesky Profile Info: Structured Bluesky data: send `handles`. Single lookup, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: single lookup · Results: data.profile Parameters: - handles (required, string, example bsky.app): The Bluesky account: a handle ("espn.com", "@espn.com"), a profile URL (https://bsky.app/profile/espn.com, any host with the /profile/ path), a DID ("did:plc:..", "did:web:..") or an at:// URI -> ONE profile record (did, handle, display_name, bio, avatar_url, banner, verified, followers, following, posts_count, likes_count, joined_at, labels, associated / verification blocks) + computed{engagement_rate, language, content_category, estimated_reach}. Single lookup: next_cursor is always null. An unknown / deactivated / taken-down account answers a status record (not_found | account_deactivated | account_takedown | invalid_input). - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - compute_metrics (optional, boolean, default true, locked in cursor): true (default) = sample the latest `feed_sample` original posts for computed.language / engagement_rate / content_category and resolve the amplifiers of the newest `reach_sample` posts for computed.estimated_reach. false = profile only (1 request; engagement_rate / estimated_reach null, language and category from the bio). - feed_sample (optional, integer, default 30, locked in cursor): Original posts sampled for the computed block (1..100, default 30). - reach_sample (optional, integer, default 5, locked in cursor): Newest sampled posts whose reposters / quoters are resolved for estimated_reach. Capped at 5 so the call fits the timeout; 0 = skip. - amplifier_pages (optional, integer, default 1, locked in cursor): Pages of 100 reposters / quoters read per sampled post before extrapolating. Capped at 1. - count_likes (optional, boolean, default true, locked in cursor): true (default) = walk the post history (100 posts a call) and sum likeCount -> likes_count (likes received). false = likes_count null. - likes_max_posts (optional, integer, default 1000, locked in cursor): Cap on the likes walk. Capped at 1000 (= 10 calls): past it likes_count is a lower bound (stats.likes_count_complete false + warning). - page_fallback (optional, boolean, default true, locked in cursor): true (default) = bsky.app og-meta partial profile when the AppView is unreachable; false = status fetch_failed. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/bluesky-profile-info/?handles=bsky.app" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/bluesky/bluesky-profile-info.md ## GET /bluesky-single-posts/ Bluesky Single Posts: Structured Bluesky data: send `urls`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.rows Parameters: - urls (required, string, example https://bsky.app/profile/bsky.app/post/3mk4lzkrnk22d): The Bluesky post: its URL (https://bsky.app/profile//post/), an at:// URI (at:///app.bsky.feed.post/) or a bare "/post/" -> 50 rows per call over the flattened thread: page 1 = the post row (row_type "post": text, author, media, like / reply / repost / quote counts, full postView, computed {engagement_rate, language, content_category, estimated_reach}, stats) + its ancestors when it is a reply (row_type "parent") + the first replies in thread order (row_type "reply", depth, thread_index, parent_uri, root_uri); later pages = replies only. A deleted / not-found / blocked post answers one status row (next_cursor null). Required on the first call; ignored when `cursor` is supplied. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - depth (optional, integer, default 6, locked in cursor): Reply levels below the post (default 6). 0 = the post alone. - parent_height (optional, integer, default 80, locked in cursor): Ancestors included as "parent" rows on page 1 when the post is a reply (default 80; 0 = none). - max_replies (optional, integer, default 0, locked in cursor): Direct replies to deliver. 0 (default) = 200 (the AppView's own cap, so the reply-index top-up only runs for threads the AppView shortened); an explicit value is capped at 500. Trimmed after reply_sort. - reply_index (optional, boolean, default true, locked in cursor): true (default) = top up direct replies past the AppView's cap from the public reply index (constellation.microcosm.blue + getPosts); false = the thread call alone. - reply_sort (optional, string, one of: api|newest|oldest|top, default api, locked in cursor): api (default - the AppView's order) | newest | oldest | top (likes, then replies), applied at every level. - compute_metrics (optional, boolean, default true, locked in cursor): true (default) = getProfile on the author (page 1) -> author_followers_count, computed.engagement_rate, flags.pinned. - estimate_reach (optional, boolean, default true, locked in cursor): true (default) = reposters + quoters follower counts (page 1) -> computed.estimated_reach. - reach_max_amplifiers (optional, integer, default 100, locked in cursor): Amplifiers resolved for estimated_reach. Capped at 100 so the call fits the timeout; past it the estimate is flagged partial. - video_duration (optional, boolean, default true, locked in cursor): true (default) = a video post gets duration_seconds from its HLS playlist (page 1). - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/bluesky-single-posts/?urls=https%3A%2F%2Fbsky.app%2Fprofile%2Fbsky.app%2Fpost%2F3mk4lzkrnk22d" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/bluesky/bluesky-single-posts.md ## GET /bluesky-user-posts/ Bluesky User Posts: Structured Bluesky data: send `handles`. Cursor-paginated, 1 credit per successful call. Cost: 1 credit per successful call · Pagination: cursor · Results: data.posts Parameters: - handles (required, string, example espn.com): Bluesky account: a handle ("espn.com", "@espn.com"), a profile URL (https://bsky.app/profile/espn.com), an at:// URI or a DID ("did:plc:x7d6j54pm22ufehkes6jo4jf") -> ONE native getAuthorFeed page (30 posts) per call in feed order, each row with the post's text, author, media, like / reply / repost / quote counts, the full post record + embed and the computed block (engagement_rate, language, content_category). An unknown / deactivated / taken-down / blocked account or an empty feed answers one status row (next_cursor null). Required on the first call; ignored when `cursor` is supplied. A raw Bluesky cursor (a v1 row's pagination.next_cursor) passed as `cursor` on the first call starts the walk there. - cursor (optional, string): Pagination cursor returned in the previous response's `next_cursor`. Pass it back unchanged to fetch the next page. When `next_cursor` is null, you have reached the last page. - filter (optional, string, one of: posts_and_author_threads|posts_no_replies|posts_with_replies|posts_with_media|posts_with_video, default posts_and_author_threads, locked in cursor): posts_and_author_threads (default) | posts_no_replies | posts_with_replies | posts_with_media | posts_with_video - getAuthorFeed's own filter (applied by Bluesky after slicing, so a page can be shorter than 30). - include_pins (optional, boolean, default true, locked in cursor): true (default) = the pinned post rides on top of page 1 (is_pinned_copy true, an extra item); false = chronological only. - compute_metrics (optional, boolean, default true, locked in cursor): true (default) = getProfile on page 1 fills account_followers_count / account_posts_count and computed.engagement_rate (carried in the cursor for later pages). - video_duration (optional, boolean, default true, locked in cursor): true (default) = video posts get duration_seconds from their HLS playlist. - raw_data (optional, boolean, default false): When true, the response carries the full raw upstream payload(s) the scraper received (GraphQL / JSON API bodies, HTML, embedded JSON-LD -- exactly as the platform returned them) under `raw_responses`, in place of the formatted items. When false (default) you get the formatted output. Pagination (`cursor` / `next_cursor`) works the same in both modes. Example: curl "https://api.scrapier.io/api/v2/bluesky-user-posts/?handles=espn.com" -H "X-API-Key: YOUR_API_KEY" Docs: https://docs.scrapier.io/api-reference/bluesky/bluesky-user-posts.md