AI Agent Hub
Back to skills
Maple Video to Article icon

Maple Video to Article

Content Creation Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_fb9b2bd3/maple-video-to-txt.

About this skill

Problem

maple-video-to-txt targets a concrete content pipeline: turn a local video or video URL into publishable illustrated Markdown. It avoids relying on image-understanding models to guess screen content, and is useful for public accounts, CSDN, summaries, and reports.

How It Works

The skill confirms input type and runtime conditions, then follows fixed scripts:
- URL input: uses scripts/video_downloader.py to download video, handling yt-dlp and filename cleanup.
- Transcription: uses scripts/video_to_text.py with faster-whisper to produce .srt and .txt.
- Frames: extracts candidate frames from the video.
- Image matching: inserts screenshots near corresponding subtitle segments by timeline, not by visual guessing.
- Writing: drafts according to user style or default templates, using only matched images.

Boundaries and Notes

It depends on Python 3.11/3.12, yt-dlp, ffmpeg, faster-whisper, and other components. First transcription may require network access to download the Whisper model; download failures often come from sandboxes, firewalls, or proxy policies. On Windows, use python rather than assuming python3 exists. For lecture, finance, or static-heavy videos, avoid over-skipping similar frames, or key visuals may be missed.

Use Cases

  • Turn a local tutorial video into a Markdown guide with matched screenshots.
  • Download an online lecture URL, transcribe it, and draft a summary article.
  • Insert extracted frames beside subtitle segments when writing a public account post.
  • Convert a finance video into a recap article using only timeline-matched images.

Best For

  • Technical content editors who need local tutorial videos turned into publishable guides.
  • Public account writers who need screenshots matched to subtitle timelines.
  • Report authors who need online videos summarized into Markdown with images.
  • Knowledge-base maintainers who want downloaded videos turned into illustrated notes.