Content Triage App
Currently running as a real personal workflow (agents, markdown files, scripts). A written explainer post covering the current setup is planned, and it's a candidate for a short walkthrough video too.
Saved-content tools tend to be either narrow and passive (read-it-later apps: capture now, decide never) or open-ended and passive (a notes app: dump it in, structure optional). Almost nothing forces an actual decision about what a saved item is for at the moment it’s saved — which feels fine short-term and produces a familiar outcome: a growing pile of unprocessed links with no signal about which ones actually matter.
Content Triage App forces that decision, one item at a time, through a swipe/card-style interface. Each saved link gets a live preview — title, description, thumbnail — and has to be assigned to exactly one bucket, no “skip for now.” Buckets map to real downstream actions (Read, Summarize, Evaluate, Discard) rather than generic tags, and every sort captures a short reason, so the record isn’t just where something went but why.
Storage is plain markdown files with frontmatter, physically moved between bucket folders on sort — which keeps the whole thing directly compatible with a personal notes vault, no export step required. The longer-term direction is per-bucket automation (agents that process a bucket’s contents on a schedule) and a self-insight layer: surfacing patterns like false-positive save rate, or which interests are actually growing versus just imagined.