2.5k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill skill-tiktok-ads-video- _meta.json479 B
- SKILL.md2.6 KB
Overview
This skill generates TikTok-style short-form ad videos by overlaying animated pill captions and mixing background audio on a base MP4. It uses pure Python (MoviePy + Pillow) and ships with built-in product presets so you don’t need a separate captions JSON file. The output is optimized for TikTok, Reels, and Shorts formats and supports custom fonts and optional music tracks.
How this skill works
You provide a base MP4 and choose a product preset and caption style; the script renders animated pill captions that fade in and out and positions them responsively over the video. Audio can be replaced or mixed with an optional music track and start time. A legacy JSON captions mode is still supported for manual timing and text control.
When to use it
- Creating product-focused TikTok, Instagram Reels, or YouTube Shorts ads quickly.
- Converting demo footage into polished social ads without Premiere or CapCut.
- Applying consistent, branded animated captions across multiple videos.
- Testing different caption styles and background audio for ad variants.
Best practices
- Start with a clean, high-frame-rate MP4 (Veo/Runway/Kling exports recommended).
- Use built-in product presets for fast, visually consistent captions.
- Supply a separate music track and set --audio-start to align beats with cuts.
- Provide custom Montserrat fonts if you need strict brand typography.
- Avoid emoji in captions; use plain text alternatives for reliable rendering.
Example use cases
- Produce a 15-second product demo for TikTok using the rain_cloud preset and subtitle_talk style.
- Batch-create ad variants by swapping --product and --style while keeping the same base video.
- Replace on-camera audio with a licensed track starting at a specified offset for synchronous transitions.
- Restore older captions workflows with a manual captions JSON for precise timing or localization.
FAQ
No. This skill is pure Python and uses MoviePy and Pillow to render animated captions and audio mixing.
Can I supply my own fonts?
Yes. Use --font-black and --font-bold to point to Montserrat files; the skill falls back to fonts in ~/.local/share/fonts/.
What if I need manual timing for captions?
The legacy captions JSON mode is still supported for precise timing and custom caption data.