Captions Overlay skill: what it does and how to install it

What it does. Classifies spoken phrases as drop, rail or embed: filler is omitted, ordinary speech stays as a verbatim lower-third, and rare peak words appear behind the subject. Treats captions as overlays on the full film, keeping compositions centred on the true frame centre instead of reserving a bottom band.

When to use it. Use for talking-head, explainer and launch-video compositions with captions. Standard mode keeps the rail; cinematic mode is not for explainer or voiceover work, where all spoken words need to remain readable.