360 Secure SkillHub
360 安全技能中心

Reel to Skill

接收 TikTok、Instagram Reel 或 X 短视频链接，调用脚本本地下载视频并提取音频，经 OpenAI 转写后分析其中的教学或工作流，最终在 skills 目录下生成简洁可复用的技能文件；若无法转写则改用抽帧并请用户提供文字稿。

olliewazza
olliewazza
热度 26.04
最近更新 2026-05-17 02:51:48
分类：开发依赖：代码执行 / 读写文件 / 第三方 API / 内部调用大模型应用场景：摘要总结 / 工作流 / 自动化脚本

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add reel-to-skill
Metadata
name
reel-to-skill
description
Download a TikTok, Instagram Reel, X video, or short-form video URL locally, extract/transcribe the audio, analyse the teaching or workflow inside it, and turn it into a concise OpenClaw skill. Use when the user sends a social video link and asks to learn from it, make a skill from it, extract the process, or build a reusable workflow without manually downloading the video.

Reel to Skill

Use this when the user provides a TikTok/Instagram Reel/X video/short-form video link and wants a skill from it.

Workflow

  1. Download the video locally with scripts/download_transcribe.py.
  2. Transcribe the audio.
  3. Extract the repeatable procedure, decision rules, examples, and caveats.
  4. Create a skill folder under skills/<skill-name>/.
  5. Keep SKILL.md concise and procedural.
  6. Do not hardcode API keys, cookies, tokens, or user secrets into the skill.

Script

Run:

python3 skills/reel-to-skill/scripts/download_transcribe.py "<url>" --out /tmp/reel-skill

The script uses:

  • yt-dlp for downloading
  • ffmpeg for audio extraction
  • OpenAI transcription if OPENAI_API_KEY is available

If transcription is unavailable, extract frames and ask the user for a transcript or use available audio tooling.