embedded-captions
๐ฏSkillfrom heygen-com/hyperframes
|Adds cinematic captions to talking-head videos using a catalog of 17+ visual identities. Supports column-flow compositing (captions behind the subject via matting and occlusion) and themed constitutions with effects like neon sign reveals and glyph-decode climaxes.
Overview
embedded-captions is a HyperFrames skill that adds captions to talking-head videos using a catalog of 17+ visual identities across three engines. Standard mode builds a clean verbatim lower-third subtitle rail with an embed climax composited into the scene behind the subject at the peak moment. Cinematic mode is pure embed with no rail, where every caption is composited behind the subject using hero typography and occlusion. Theme mode provides complete themed constitutions with body paradigm, hero setpiece, front effects, and plate reactions from named identities like ordnance, terminal, neonsign, stardust, and stomp.
The pipeline is deterministic: decision gate and identity selection from CATALOG.md, project initialization, parallel preparation (matte extraction, transcription, audio envelope analysis, and safe-zone computation), creative choices authoring, and final composition. The skill routes by identity rather than by mode, keeping selection simple.
Key Features
- 17+ visual identities in a single catalog, covering clean subtitles to VFX-grade cinematic effects
- Three compositing engines: Standard (rail + embed climax), Cinematic (pure embed with occlusion), and Theme (full themed constitutions)
- Automated matting pipeline using rembg for subject extraction, enabling captions composited behind the subject
- Parallel preparation: matte extraction, Whisper transcription, audio envelope analysis, and safe-zone computation run simultaneously
- Named theme identities (ordnance, terminal, neonsign, stardust, stomp, and more) each with unique visual paradigms
- Deterministic pipeline with computed timings and compiled compositions rather than hand-written HTML
Who is this for?
- Video creators who want professional-grade captions on talking-head content without After Effects
- YouTubers and content creators looking for cinematic caption styles beyond standard subtitles
- Social media creators who need eye-catching VFX-style captions for short-form content
- Production teams seeking automated, repeatable caption workflows with consistent visual quality
Same repository
heygen-com/hyperframes(48 items)
Installation
npx vibeindex add heygen-com/hyperframes --skill embedded-captionsnpx skills add heygen-com/hyperframes --skill embedded-captions~/.claude/skills/embedded-captions/SKILL.mdSKILL.md
More from this repository10
CLI tool for the Hyperframes video rendering framework that provides commands to create, preview, lint, and render HTML-based video compositions to MP4.
Skill for authoring HTML-based video compositions with the Hyperframes framework, covering captions, TTS, audio-reactive animation, and transitions for AI agent-driven video creation.
Skill for installing blocks and components into Hyperframes projects via the `hyperframes add` command, providing access to 50+ ready-to-use overlays, shader transitions, and data visualizations.
A skill for the HyperFrames open-source video rendering framework that teaches AI agents to write correct HTML-based video compositions and GSAP animations, enabling them to render video from HTML code.
The foundational technical contract for HyperFrames HTML video compositions, defining data attributes, clip/track structure, sub-compositions, variables, media playback, deterministic rendering rules, and project validation.
Comprehensive animation skill for HyperFrames providing atomic motion rules, multi-phase scene blueprints, scene transitions, motion-design techniques, and seven runtime adapters including GSAP, Lottie, Three.js, Anime.js, CSS keyframes, Web Animations API, and TypeGPU.
Non-animation creative direction for HyperFrames videos covering design specs, color palettes, typography, narration scripting, beat planning, audio-reactive visuals, composition patterns, and brand/style decisions.
Resolves media needs (background music, sound effects, images, icons) into frozen local files by searching the HeyGen catalog, downloading the best match, and registering it in a manifest. One resolve command handles the full cascade from cache lookup to provider search to file freezing.
A media framework by HeyGen for building interactive video experiences with composable frames and real-time rendering capabilities.
A fallback HyperFrames composition authoring skill for custom video creation when no specialized workflow fits. Handles multi-scene pieces, brand reels, montages, title cards, motion posters, static loops, and freeform compositions at any length or format.