chengyi-ai/native-subtitle-quote-image

What it solves

It automates the creation of high-quality 3:4 aspect ratio "quote images" from videos. It solves the problem of manually capturing frames and arranging subtitles for social media posts, ensuring that the resulting images look professional and maintain a consistent layout regardless of the number of subtitles used.

How it works

The tool processes local video files or YouTube links to extract specific frames and subtitles. It operates in two distinct modes:

• Native Mode: Crops subtitles that are already "burned-in" to the video pixels, avoiding OCR redraws or translations.
• Script Mode: Overlays verified text from a JSON script onto real video frames, clearly marking them as post-production subtitles.

It can be used as a standalone Python script or integrated as a "Skill" for AI agents (like Claude Code or Codex), allowing users to generate these images using natural language commands.

Who it’s for

• Content creators who want to turn video highlights into shareable long-form images.
• Social media managers needing a fast way to generate visual quotes from video content.
• Users of AI agents who want to automate the visual asset production pipeline from a video transcript.

Highlights

• Dual Subtitle Modes: Strictly separates native burned-in subtitles from script-based overlays.
• Agent Integration: Compatible with open Agent Skills formats for one-sentence execution via AI agents.
• Dynamic Layout: Automatically adjusts the ratio between the main image and subtitle bars to maintain a compact 3:4 look.
• YouTube Support: Integrates with yt-dlp to handle online videos and metadata.