graphic-overlays
Package an existing talking-head / interview / podcast video by layering timed
by heygen-comFree
Description
graphic-overlays is an agent skill from HeyGen's HyperFrames project that packages an existing talking-head, interview or podcast video by layering timed, designed graphic cards on top of it. The video plays in full, and the cards are timed to what is being said: titles, lower-thirds, data callouts, quotes, side panels and picture-in-picture. The agent designs each card's timing and content, writes the card HTML in the conversation, assembles a single composition, and renders it to MP4 with HyperFrames. There is no fixed list of card types, so the overlays come from what the transcript says. The skill sets a routing rule first. If the user wants plain captions or subtitles, the request goes to a captions skill, and if it is a single short element, such as one logo sting or lower-third, it goes to a motion graphics skill. The clip itself plays untouched, so re-timing, recoloring, reframing, reordering and audio editing are out of scope and belong to an editor. A request to build a video from a URL, a topic or a pull request goes to the creation workflows. It is a graphic-packaging sibling of the embedded captions skill, which adds the spoken words as a readable subtitle. It suits creators who have a recording and want on-screen graphics added to it. It passes Gen Agent Trust Hub, Socket and Snyk on skills.sh, and has about 29,100 installs. Install it with npx skills add https://github.com/heygen-com/hyperframes --skill graphic-overlays.…
Highlights
- Layers timed graphic cards onto an existing talking-head or interview video
- Titles, lower-thirds, data callouts, quotes, side panels and picture-in-picture
- Card timing and content follow what the transcript says
- Routes plain captions and single elements to other skills
Ratings & Reviews
No ratings yet
Be the first to rate graphic-overlays.
People also visit
podcast-generation
The user requests to generate, create, or produce podcasts from text content
graphic-overlays FAQ
What does it add to my video?
Timed graphic cards such as titles, lower-thirds, data callouts, quotes, side panels and picture-in-picture, matched to the speech.
Does it edit the clip?
No. The clip plays untouched. Re-timing, recoloring, reframing and audio editing are out of scope.
I only want subtitles. Is this the right skill?
No. The skill routes plain captions to a separate captions skill.
How is the result made?
The agent writes the card HTML, assembles one composition and renders it to MP4 with HyperFrames.
How do I install it?
Run npx skills add https://github.com/heygen-com/hyperframes --skill graphic-overlays.
What are the best graphic-overlays alternatives?
Popular alternatives to graphic-overlays include slideshow, podcast-generation and figma. See all graphic-overlays alternatives