Community · E2 · artifact verified
Triage YouTube comments by reply worthiness
Paste a YouTube URL and every comment in the section is classified on four typed axes - then ranked so the ones actually worth a reply float to the top, for creators drowning in comment volume.
01 · Role in the system
What Jev does here
The triage runs per comment: one pass asks four typed questions over the text - what kind of comment it is, whether it carries a question, its tone, whether it needs a reply at all - and code ranks the section from the calibrated answers, floating the answerable questions and genuine feedback above noise. A smoke test exercises the pipeline end to end, and the creator-facing output is a ranked list rather than a filter: nothing is hidden, the reply queue is simply ordered by what deserves attention first.
02 · Control boundary
Where Jev sits
Jev as the four-axis judge of a creator's comment queue: one typed pass per comment (kind, question, tone, needs-reply), code ranks on the calibrated answers, and the creator works down a sorted queue instead of an unfiltered wall.
Code owns the loop, permissions, thresholds, validation, and side effects. Jev owns only the bounded judgments described above.
03 · Known limits
What this evidence does not prove
- Single video per run; no channel-wide batch mode.
- Fresh project (0 stars); the four axes and thresholds are the author's choices.
- Comment language coverage untested beyond English.
04 · Attribution
Public sources
This is a Community record: the project was published by a third-party community author.
- anshparashar-10 ↗Community · github · public · checked 2026-10-07