Hyperframes Captions
Add embedded captions to talking-head video using the HyperFrames caption composition system, with rail subtitles and embed climax words composited behind the subject using matting.
Install
npx skills add https://github.com/heygen-com/hyperframes --skill hyperframes-captionsWhat is this skill?
- Rail + embed caption model: lower-third subtitles and behind-subject climax words
- Three render engines: Cinematic, Standard, Theme modes
- Matte-based subject segmentation (u2net_human_seg)
- Preview composites ~2s/frame before full render
- 17-identity catalog for visual style selection
Adoption & trust: 46 installs on skills.sh.
Who is it for?
Creators adding stylized embedded captions to talking-head videos using HyperFrames
Skip if: Adding captions to multi-speaker footage or non-HyperFrames video pipelines
What do I get? / Deliverables
- captioned video file
- caption composition plan.json
- preview frames
Recommended Skills
Video Editagentspace-so/runcomfy-agent-skills
Image To Videoagentspace-so/runcomfy-agent-skills
Image Editagentspace-so/runcomfy-agent-skills
Flux Kontextagentspace-so/runcomfy-agent-skills
Nano Banana 2agentspace-so/runcomfy-agent-skills
Nano Banana Editagentspace-so/runcomfy-agent-skills
Journey fit
Primary fit
Adding captions to video content is a build-phase creative production step: you author the caption composition plan, preview frames, and render the final captioned video. Caption composition in HyperFrames is a frontend/UI task: it involves CSS-level layout of caption layers, visual QA of contrast and occlusion, and HTML composition authoring.