Our data. Our media. No Instagram.

Every scrape is written to our own database, and every image is copied into our own storage while the source link is still alive. Analytics, decks and AI matching all run off this — nothing here breaks when an Instagram CDN link expires.

Rows stored

—

across all tables

Media cached

—

Still on Instagram

—

images not yet copied over

Copy remaining Instagram media into our storage

Runs in small batches. Links that already expired can't be recovered — those keep their original URL and are re-cached on the next re-scrape.

Take the data with you

Read-only exports of everything we've collected. CSV opens in Sheets/Excel; JSON is the same payload the product and the MCP agents read.

Table	Rows	Download

API

GET /api/public/export?table=creators&format=json&limit=1000
GET /api/public/export?table=international_posts&format=csv
GET /api/public/media?key=<storage_key from media_assets>

Each talent row also carries raw_profile — the complete untouched scrape — so you can re-run any analysis later without touching Instagram again.

Start fresh

Delete all scraped data

Wipes every talent profile, brand, collaboration, generated deck and cached image. Use this before adding a new roster so everything is re-scraped with the latest logic. This can't be undone.