GH Repository · chengyi-ai
native-subtitle-quote-image
保留视频内嵌字幕，精确取帧并生成 3:4 社交长图的 Agent Skill
- stars
- 700
- 30-day movement
- +472157/day
- Related entries
- 61
- Connections
- 1
skill-packPythonsubtitlessocial-mediacodexvideo-processingagent-skills
An Agent Skill that extracts frames from videos while preserving the embedded (native) subtitles, then composes them into 3:4 portrait long images for social media. It ships as a skill pack (Python, MIT) for use with agent tooling such as Codex.
Use it when you want quote-style screenshots from videos that keep the original burned-in subtitles, without manual frame picking and layout work.
Use it to
- Extract subtitle-bearing frames from a video
- Generate 3:4 social media long images
- Package the skill into an agent workflow
- Automate video-to-quote-image pipelines
For Agent developers and social media content creators
- Role
- skill-pack
- Language
- Python
- Licence
- MIT
- Forks
- 96
- Open issues
- 2
- Last push
- 2026-08-24
- Latest release
- v1.0.0 · 2026-08-16
- Skills shipped
- 1
topicsagent-skillscodexsocial-mediasubtitlesvideo-processingpython