skills.rest
CodeWithSally
CodeWithSallyOFFICIAL
Explore all skills by this author →
GitHub

captions-overlay

Defines caption overlay doctrine for compositing subtitles on video compositions without reserved bands.

13|17|Updated Jan 30, 2025
Search Tags:#layout#subtitles#captions#overlay#talking-head#video composition#lower-third
One-click install
npx skills add https://github.com/CodeWithSally/CodeWithSally-Apex-English --skill captions-overlay-codewithsally
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: captions-overlay
Source: https://github.com/CodeWithSally/CodeWithSally-Apex-English/tree/main/Sessions/ClaudeCode/Session007/.agents/skills/captions-overlay
Command: npx skills add https://github.com/CodeWithSally/CodeWithSally-Apex-English --skill captions-overlay-codewithsally

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When adding captions to talking-head or launch videos, creators often misplace text by embedding every word or shifting the whole composition up to reserve a dead bottom band for subtitles. This Skill provides the doctrine that prevents both mistakes. ## Core Features & Use Cases - Caption Model (drop / rail / embed): Classifies every spoken phrase as dropped filler, verbatim lower-third rail text, or a scarce embedded peak word composited behind the subject. - Overlay Law: Enforces that captions are an overlay composited on top of the film, so compositions stay centered on the true vertical center (y = H/2) with no keep-out band. - Use Case: When laying out a product launch video that will carry captions, apply this doctrine to center content at the true frame center, let imagery run full-bleed to the canvas bottom, and promote only one headline word per beat to an embed. ## Quick Start Apply the captions-overlay doctrine to lay out my launch video composition with captions enabled, keeping the content centered on the true frame center.

Frequently Asked Questions about captions-overlay

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add captions to a talking-head video without ruining the layout?▼

Treat captions as an overlay composited on top of the film, not a reserved zone. Keep the composition centered on the true vertical center (y = H/2), let content run full-bleed to the canvas bottom, and place the verbatim text on a lower-third rail.

What is the drop, rail, embed caption model?▼

Every spoken phrase is classified as drop (filler like ums, not shown), rail (ordinary verbatim content on a lower-third subtitle), or embed (a scarce peak word composited behind the subject). The rail carries most text; embeds are limited to one per beat.

Should I reserve a bottom band in my video layout for subtitles?▼

No. Reserving a keep-out band and shifting content up is explicitly the bug, not the fix. Captions are an overlay layer, so center the composition on the true frame center and only avoid parking critical small readable text in the bottom center span.

When should I use embed-style captions instead of a subtitle rail?▼

Use embeds only for genuine peak moments, at most one per beat, never two adjacent or co-visible. Cinematic mode, which drops the rail entirely, is only for pure-cinematic asks, never for explainers or voiceover where words must read.

Does this Skill replace the embedded-captions skill?▼

No, it supplements the upstream embedded-captions skill and applies on top of it. It quotes the rail-plus-embed model from embedded-captions and adds the overlay law from the product-launch-video scene agent's constraint #13.

Related Skills

Explore More Design & Creative Skills→
chuanyue98chuanyue98

space-video-edit

Cuts talking-head videos by removing filler words and mistakes, then assembles final MP4 with subtitles.

Community
Advanced
heygen-comheygen-com
43.5k

captions-overlay

Applies caption overlay doctrine for compositing subtitles on video compositions.

Official
Intermediate
heygen-comheygen-com
43.5k

embedded-captions

Adds verbatim or cinematic captions to talking-head videos using transcription and subject matting.

Official
Advanced
neighbor-game-aineighbor-game-ai

bria-rmbg

Remove image backgrounds automatically using BRIA RMBG 2.0 via the Replicate API.

Official
Intermediate