talking-head-recut

What problem does it solve?

This Skill eliminates the tedious, time-consuming work of manually designing, timing, and syncing graphic overlays (titles, lower-thirds, data callouts, quotes, side panels, picture-in-picture) to existing talking-head, interview, or podcast videos, ensuring overlays align perfectly with spoken content without requiring manual video editing or NLE tool use.

Core Features & Use Cases

• Transcript-synced graphic cards: Automatically generates timed overlay cards synced to local Whisper transcriptions of the video's audio, with no third-party API keys or rate limits required.
• Customizable visual design: Offers 10 style presets, 4 layout options, and 3 video frame styles to match the video's tone, from academic and editorial to experimental and social media-focused.
• Use case: Perfect for content creators, podcasters, interview producers, and social media managers who want to add professional, designed on-screen graphics to pre-recorded talking-head footage to increase engagement and clarity.

Quick Start

Use the talking-head-recut skill to add synced graphic overlays to my local talking-head video file at /path/to/my-interview.mp4, using the default auto card count and 16:9 landscape output.