I’ve been exploring how small, well-placed cues might help people notice reasoning flaws—without disrupting conversation or turning reflection into policing.

FallacyTag is an early prototype of that idea:

• Tags likely fallacies in real-world text (demo: https://coherentdrift.github.io/fallacytag/demo/?theme=acade...)

• 30s video walkthrough included

• Designed to support reflection in writing, media, even conversation—not just detection

• Full write-up and system proposal here: https://coherentdrift.github.io/fallacytag/pages/04-introduc...