faceless-explainer

Paste the article text and the skill saves it verbatim, drafts a storyboard and narration script, generates voice and background music, builds one HTML composition per frame, and renders a final MP4. You approve the plan and the final preview at gated review steps.

A faceless explainer is a video where every visual is invented rather than captured from a website or footage: typography, abstract graphics, diagrams, and data visualizations. This skill builds those visuals per scene in HyperFrames instead of recording a product or site.

Yes. Add brand colors and fonts to capture/extracted/tokens.json before Step 2, and the build-frame script remixes a chosen preset onto those tokens, mapping colors by role and swapping display and body font families automatically.

Yes. The audio step generates TTS narration with word timings via HeyGen when signed in or Kokoro offline, retrieves BGM matching the storyboard's music mood, and fetches per-frame SFX. Setting music to none with no script marks the project fully silent.

Do not use it for videos built from a website or product, such as promos or site tours; those route to the product-launch-video skill instead. Any unclear intent should go through the hyperframes intent layer first.