Introduction
Overview of the Instagram Saved Posts Tracker, its features, and how it safeguards your visual bookmarks.
Welcome to Saved Posts Tracker
Saved Posts Tracker is a production-grade, self-hosted web application designed to automatically synchronize, mirror, and intelligently index your saved Instagram bookmarks into a database and media storage that you control.
+--------------------------------------------------------------------+
| Saved Posts Tracker |
| |
| +----------------+ +-------------------+ +-----------------+ |
| | Multi-Profile |-->| Resumable Scraper |-->| Cloudinary Sync | |
| +----------------+ +-------------------+ +-----------------+ |
| | |
| +----------------+ +-------------------+ v |
| | Event Timeline |<--| Qdrant Vector AI |<-- [ High-Res CDN ] |
| +----------------+ +-------------------+ |
+--------------------------------------------------------------------+Why We Built This
Instagram saved collections suffer from several critical shortcomings:
- Expiring Media Signatures: Direct Instagram CDN URLs expire after a short time.
- Account Deletions & Renames: When creators change their handle or get banned, links break permanently.
- Zero Semantic Search: Native Instagram offers no visual concept or facial recognition search.
- Single-Account Friction: Switching between research and personal accounts requires continuous logout/login cycles.
Saved Posts Tracker addresses every one of these problems with a unified, self-hosted architecture.
Key Capabilities
- Netflix-Style Multi-Profile Architecture: Connect multiple Instagram accounts with isolated cookies, custom user-agents, avatars, and separate datasets.
- Resumable Checkpoint Scraper: Uses
checkpointMaxIdto safely pause and resume scraping across rate limits and network interruptions without duplicates. - Append-Only Event Timelines: Detects and logs username renames (
from @old to @new), account deletions (lost), recoveries, and privacy flips. - Cloudinary CDN Mirroring: Automatically syncs photos, multi-slide carousels, and video clips to permanent cloud storage.
- In-Browser & Qdrant Vector AI (Coming Soon): 512-dimensional CLIP multimodal search and TensorFlow.js in-browser biometric face detection.
Next Steps
- Check out the Quickstart Guide to deploy in 5 minutes.
- Learn about the System Architecture.