skill-omni-creation

Generates standard Skill Markdown files from web pages or video tutorials.

7.1k|1.1k|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/openJiuwen-ai/jiuwenswarm --skill skill-omni-creation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-omni-creation
Source: https://github.com/openJiuwen-ai/jiuwenswarm/tree/main/jiuwenswarm/resources/agent/workspace/skills/skill-omni-creation
Command: npx skills add https://github.com/openJiuwen-ai/jiuwenswarm --skill skill-omni-creation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, beautifulsoup4, Pillow, requests, yt-dlp, and includes scripts (resource) components.

What problem does it solve?

Turning an online tutorial, documentation page, or instructional video into a reusable, well-structured Skill file normally requires manual reading, screenshot curation, and formatting. This Skill automates that entire pipeline: it scrapes the page or extracts video frames, filters relevant images, and writes a standards-compliant SKILL.md with optional verified helper scripts.

Core Features & Use Cases

  • Web-to-Skill Conversion: Scrapes tutorial pages with Playwright, downloads and deduplicates images, and produces a structured SKILL.md with steps and screenshots saved under references/.
  • Video-to-Skill Extraction: Downloads videos from Bilibili, YouTube, Vimeo, or Xiaohongshu, extracts up to 90 frames with ffmpeg, and converts key frames plus steps into a Skill file.
  • Automatic Script Generation: Detects when a task requires code, then writes, verifies, and finalizes companion Python scripts in scripts/ before generating the final Skill.
  • Use Case: A user says "turn this Bilibili video tutorial into a Skill" — the pipeline downloads the video, samples frames in batches, selects key screenshots, and writes a complete SKILL.md with image references.

Quick Start

Generate a Skill from this tutorial URL and save it with its images and any needed scripts.

Frequently Asked Questions about skill-omni-creation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a Skill file from a web page or URL?

Provide the tutorial URL and the pipeline scrapes the page with Playwright, downloads and reviews images, then writes a standard SKILL.md with steps and image references. The scrape_page.py script handles content extraction and produces staged JSON for downstream processing.

How to convert a video tutorial into structured documentation?

The analyze_video.py script downloads videos via yt-dlp or the Bilibili API, extracts up to 90 frames with ffmpeg at adaptive rates, and presents them in batches of 5 for review. Selected frames are saved as references and embedded into the generated Skill Markdown.

What video platforms are supported for frame extraction?

Bilibili, YouTube, Vimeo, and Xiaohongshu links are supported. Bilibili uses a public API with WBI signing and resumable downloads, while other platforms go through yt-dlp with browser cookie fallbacks.

What dependencies are required to run the web scraping pipeline?

The pipeline requires Python 3.11+ with playwright, beautifulsoup4, Pillow, and requests, plus a working Playwright Chromium browser. Video processing additionally needs ffmpeg, ffprobe, and yt-dlp installed in the same Python environment.

Why does the scraper stop with ENVIRONMENT_BLOCKED?

ENVIRONMENT_BLOCKED means the environment gate could not install missing Python packages or launch Chromium, often due to missing Linux system libraries or lack of sudo privileges. The pipeline halts by design; install the reported dependencies manually and rerun.

When does the pipeline generate helper scripts for a Skill?

Scripts are generated only when the content describes tasks requiring code, such as batch processing, API calls, or multi-command CLI workflows. Pure GUI click-through tutorials skip script generation and produce a text-plus-images Skill only.