Agent SkillsIllyism/transcribe-cli › video-to-social

video-to-social

GitHub

将视频转录文本转换为适用于X、LinkedIn或通讯的发布就绪内容,包括清理口语废话、设计钩子和遵循平台格式限制。

skills/video-to-social/SKILL.md Illyism/transcribe-cli

Trigger Scenarios

用户要求将视频或转录内容转换为社交媒体帖子 用户请求生成Twitter线程、LinkedIn文章或新闻通讯草稿

Install

npx skills add Illyism/transcribe-cli --skill video-to-social -g -y
More Options

Use without installing

npx skills use Illyism/transcribe-cli@video-to-social

指定 Agent (Claude Code)

npx skills add Illyism/transcribe-cli --skill video-to-social -a claude-code -g -y

安装 repo 全部 skill

npx skills add Illyism/transcribe-cli --all -g -y

预览 repo 内 skill

npx skills add Illyism/transcribe-cli --list

SKILL.md

Frontmatter
{
    "name": "video-to-social",
    "description": "Convert SRT transcripts or video transcriptions into publish-ready X\/Twitter threads, LinkedIn posts, or newsletter drafts. Use whenever the user asks to turn a video or transcript into a social post, X thread, or LinkedIn article."
}

Convert Video Transcript to Social Content

Turn raw video transcriptions from @illyism/transcribe into ready-to-post social media content.

Platform Decision Procedure

Walk this flowchart to format content for the target platform:

Which platform is the content intended for?
├── X / Twitter Thread
│   ├── Tweet 1 (Hook): Standalone strong claim or counter-intuitive stat from the video. Max 280 chars. Must end with "🧵👇"
│   ├── Tweets 2-6 (Body): Single core idea per tweet. Max 280 chars each. Use bullet points and line breaks.
│   └── Final Tweet (Call to Action): Link to original video or invite discussion.
├── LinkedIn Post
│   ├── Line 1 (Hook): High-friction, single line that forces a "see more" click.
│   ├── Paragraphs 2-4: Short 1-2 sentence paragraphs with whitespace between them.
│   ├── Key Takeaways: 3-5 bullet points using "• " or emoji anchors.
│   └── Bottom: 3 relevant hashtags (#VideoMarketing #ContentCreation).
└── Newsletter / Article Draft
    ├── Title: Punchy, action-oriented headline.
    ├── Introduction: Context from video thesis.
    ├── Subheadings (H2): Structured by key topic timestamps.
    └── Conclusion: Final key takeaway + next steps.

Content Rules

  1. Convert Speech to Written Style: Spoken dialogue contains filler words ("like", "um", "you know"), repetition, and incomplete sentences. Remove all filler while preserving the original voice and argument.
  2. Hook Integrity: Extract the most surprising or controversial insight from the first 30% of the video to form the opening hook.
  3. Strict Length Limits:
    • X / Twitter: Every individual post MUST be strictly ≤ 280 characters.
    • LinkedIn: Keep total length under 3,000 characters; maximize readability with whitespace every 2 lines.

Example Output (X / Twitter Thread)

1/ Hooks don't matter as much as you think. 

Pull up your worst-performing video and look at the watch time graph. 

People STILL watch the first 3-4 seconds. 

Here are 3 things that ACTUALLY drive retention 🧵👇

2/ 1. The 5-second cliff

The real drop-off happens between seconds 5 and 15. 

If you don't deliver on the implicit promise by second 5, they swipe away.

3/ 2. Visual reset frequency

Change angles, text overlays, or graphics every 2.5 seconds. 

Audio keeps attention, but visual movement prevents scrolling.

Version History

  • 0983893 Current 2026-08-20 15:23

Same Skill Collection

skills/format-converter/SKILL.md
skills/summarize-transcript/SKILL.md
skills/transcribe/SKILL.md
skills/youtube-chapters/SKILL.md

Metadata

Files
0
Version
d71b75c
Hash
c7376817
Indexed
2026-08-20 15:23

Accueil - Wiki
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-29 21:48
浙ICP备14020137号-1 $Carte des visiteurs$