youtube-instant-article

What This Skill Does

The youtube-instant-article skill is an advanced automation tool for OpenClaw designed to bridge the gap between long-form video content and readable, professional articles. By leveraging OpenAI's GPT-5.2 for semantic analysis and the Telegraph API for hosting, the skill automatically parses YouTube URLs to extract titles, identify key moments, generate timestamped summaries, and capture relevant visual slides. It then composes these elements into a beautiful, formatted document that is compatible with Telegram's Instant View feature, allowing users to consume video content through a reader-friendly interface instead of watching lengthy clips.

Installation

To integrate this skill, ensure you have the required environment variables (TELEGRAPH_TOKEN and OPENAI_API_KEY) configured in your /Users/viticci/clawd/.env file. Install the skill by running the following command in your terminal: clawhub install openclaw/skills/skills/viticci/youtube-instant-article

Ensure your local environment includes the necessary dependencies: summarize (v0.10.0+), jq, and curl. The script is optimized for zsh to ensure compatibility with associative array handling.

Use Cases

• Converting lengthy tech reviews or tutorials into referenceable study notes.
• Extracting key visual slides from presentation videos to share with colleagues.
• Creating "gist" summaries for research, saving time on video playback.
• Building personal archives of educational content using the Telegraph platform.

Example Prompts

1. "Summarize this video for me: https://youtube.com/watch?v=example123"
2. "Can you turn this tutorial into a Telegraph article so I can read the key points? https://youtu.be/abc456"
3. "Give me the gist of this video and include the most important slides: https://youtube.com/watch?v=xyz789"

Tips & Limitations

• Wait for Instant View: Telegram requires 1-2 minutes to register new Telegraph pages; if the ⚡ button isn't immediate, wait a moment and refresh.
• Slide Limit: Use the --slides-max flag to control the density of images in your article; the default is 6.
• Processing Time: The skill is highly efficient, usually completing within 30-60 seconds, though this depends on the duration of the video and the depth of the analysis required by the LLM.
• Automation First: Never manually generate articles; the generate.sh script handles essential formatting and header structure required by the Instant View engine.