A camera that follows the face, big yellow captions, and cuts with no breath in them.
What it made0:00 spent editing this videoA 74-second reel of mine, edited with this skill from one phone take.Voice only here. The posted reel also has music I can't share.
Drop in a raw vertical talking-head clip and Claude cuts every pause, runs a virtual camera that follows the face and punches in on the key word, adds small and big captions, colour flashes, short b-roll on the right words and a glass call-to-action card. Built in code with HyperFrames, with a measured -14 LUFS master.
Raw in
One raw vertical talking-head video (phone footage is fine) and, optionally, b-roll you own.
Finished out
A 1080x1920 reel, typically 20-60 seconds: tight cuts, face-following camera, two caption tiers, flashes, short b-roll, a call-to-action card, sound effects and a -14 LUFS mix. A v2 mode adds an always-moving operator camera, 3D captions and floating b-roll cards inside the Instagram safe area.
The look
breath-tight cuts, every pause squeezed
virtual camera follows the face and punches in on the key word
small white 2-3 word captions that hard-cut
big two-line captions: white line plus a yellow-gradient line of the same width
colour edge flashes on the beats
0.5-1 second b-roll inserts triggered by words
frosted-glass call-to-action card
varied musical sound effects, never one repeated pop
The first thing you paste
Use the face-follow-edit skill. Raw footage: ./raw/take.mov (vertical, one person talking, language: English).
Make the BT.709 proxy and the cleaned voice, show me the phrases with their gaps and pick the cleanest complete take per beat.
Cut it breath-tight, align the exact script, then write the plan: big captions on the payoff of each beat, flashes, b-roll
only from ./raw/broll/ on the words that name things, and a glass call-to-action card on the last line.
Build it, run check and snapshots, and give me a contact sheet plus the pre-render checklist. Do not render until I say go.