Guides
How to scrape and access social media data with APIs
Evergreen tutorials with runnable examples — when to build it yourself, when to call an API, and how to get real JSON back in minutes.
Where to start
The four most recently updated guides — or browse everything below and filter by platform.
Cross-Platform Creator Profiles with One API (2026)
Look up the same creator on TikTok, YouTube, Instagram, and X with one schema and one API key — no per-platform scrapers, no drifting field names.
How to Pull Facebook Group Posts with an API (2026)
List posts from a public Facebook group as paginated JSON — sorted feeds, cursor pagination, comment threads, and local intel workflows you can run on a cron job.
How to Get YouTube, TikTok & Instagram Transcripts with One API (2026)
Pull spoken text from public YouTube, TikTok, and Instagram Reels as clean JSON — caption tracks, auto-generated speech, optional AI fallback for uncaptioned TikToks, and patterns for batch jobs and RAG pipelines.
Reddit Product Research with an API (2026)
Search public Reddit posts for competitor mentions, switching intent, and feature requests — normalized JSON you can pipe into spreadsheets, dashboards, or AI summaries.
All guides
Filter by platform, or browse everything. Each tutorial uses the same auth and response envelope as production API traffic.
Cross-Platform Creator Profiles with One API (2026)
Look up the same creator on TikTok, YouTube, Instagram, and X with one schema and one API key — no per-platform scrapers, no drifting field names.
Updated
How to Pull Facebook Group Posts with an API (2026)
List posts from a public Facebook group as paginated JSON — sorted feeds, cursor pagination, comment threads, and local intel workflows you can run on a cron job.
Updated
How to Get YouTube, TikTok & Instagram Transcripts with One API (2026)
Pull spoken text from public YouTube, TikTok, and Instagram Reels as clean JSON — caption tracks, auto-generated speech, optional AI fallback for uncaptioned TikToks, and patterns for batch jobs and RAG pipelines.
Updated
Reddit Product Research with an API (2026)
Search public Reddit posts for competitor mentions, switching intent, and feature requests — normalized JSON you can pipe into spreadsheets, dashboards, or AI summaries.
Updated
Social Comments for Sentiment Analysis with an API (2026)
Pull paginated comments from TikTok, YouTube, Instagram, Reddit, and more as normalized JSON — ready for keyword rules, classifiers, or LLM batch jobs.
Updated
Social Fetch vs DIY Scraping — Honest Tradeoffs (2026)
When to run your own headless browsers, when to call an API, and how to compare true cost — maintenance time, breakage, schema drift, and credit metering side by side.
Updated
How to Use Social Fetch with Cursor & Claude (MCP + llms.txt) (2026)
Connect Cursor, VS Code, or Claude to Social Fetch via MCP — OAuth tools for every endpoint, llms.txt for agent orientation, and the same credit billing as the REST API.
Updated
Build a Social Listening Dashboard with One API (2026)
Watch lists, cron polling, dedupe, spike detection, and Slack alerts — snapshot brand mentions across TikTok, X, YouTube, and Reddit with one JSON schema.
Updated
How to Get a TikTok Video Transcript with an API (2026)
TikTok has no transcript export, and most videos have no captions at all. Here's how to pull the spoken text from any public TikTok as clean JSON — one GET request, an AI fallback for uncaptioned videos, and timestamped WebVTT you can reuse.
Updated
How to Scrape TikTok Data with an API (2026)
Three honest ways to get public TikTok data into your product — DIY browser scraping, TikTok's official APIs, or one GET request that returns normalized JSON. With the real tradeoffs and runnable code for each.
Updated
About these guides
These guides are written for engineers who already know what they want — a transcript, a comment thread, a keyword search result — and need the shortest path from curl to something their pipeline can store. Each one names the Social Fetch routes we would actually call, shows the response fields that matter, and admits where DIY scraping still wins on cost or control.
We group tutorials by job, not by platform logo. Listening guides cover cron polling and dedupe. Research guides cover Reddit threads and Facebook groups where the big networks' search APIs miss the conversation. Creator guides cover cross-platform profile lookups before you approve spend. Agent guides cover MCP setup in Cursor and Claude when you want tools instead of copy-pasted curl.
Most pages include a "build vs buy" section with real maintenance math — proxy bills, DOM breakage, schema drift — because pretending scrapers are free after the first deploy helps nobody. If you want vendor comparisons without implementation detail, use the compare hub. If you want workflow pages that map jobs to endpoints without platform walkthroughs, use the use cases index.
Every guide uses the same response envelope as production traffic: data, meta.creditsCharged, meta.requestId, and typed errors your code can branch on. When a tutorial shows pagination, it shows lookupStatus and cursor fields the way search routes actually behave — not a simplified fake response that breaks the first time you ship.
Platform filters below are a convenience, not the organizing principle. A TikTok scraping guide and a cross-platform creator guide may both matter for the same sprint if you are vetting influencers and archiving their recent posts. Start from the "where to start" list if you are not sure which file to open first.
Guides are updated when endpoints change shape or when we learn something expensive in support — like a route that silently omits fields on private accounts. Check lastUpdated on each card. For shipping announcements and post-mortems written for a wider audience, the blog covers reliability and API design without step-by-step curl.
Same API underneath every guide
Auth, meta.requestId, meta.creditsCharged, and error codes look the same whether you copy curl from a tutorial or call routes from your backend. New accounts get 100 free credits with no card.